Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2010-05-17

00:05:49 Quit efyx (Remote host closed the connection)
00:10:59 Quit DerPapst (Quit: Leaving.)
00:12:49 Part max242
00:14:43saratogai like the pictureflow hotkey patch, its very handy
00:14:58saratogawho should I bug about getting it reviewed?
00:15:36 Quit stripwax (Quit:
00:17:31 Join Kitr88 [0] (
00:20:38 Quit Kitar|st (Ping timeout: 268 seconds)
00:21:43 Quit Kitr88 (Ping timeout: 246 seconds)
00:22:43 Quit domonoky (Read error: Connection reset by peer)
00:23:00 Join shaggy-h [0] (
00:27:59 Join Kitar|st [0] (
00:37:56 Quit bieber (Quit: Ex-Chat)
00:41:41 Quit ender` (Quit: If everything seems to be coming your way, you're probably in the wrong lane. -- Borstelmann's Rule)
00:50:55 Join Ebert [0] (
00:54:49 Quit pamaury (Ping timeout: 252 seconds)
01:01:06Ebertmy gigabeat F's hard drive keeps churning away for at least 5 minutes on every startup
01:01:34Ebertive tried some new builds, but even on the latest stable i just put on, it does the same
01:01:46 Quit hebz0rl (Quit: Ex-Chat)
01:02:36EbertI guess it could be anything, but if someone has a suggestion for a settings adjustment please let me know
01:03:20 Join bieber [0] (
01:07:25 Join solexx [0] (
01:09:56 Quit solexx_ (Ping timeout: 248 seconds)
01:12:08pixelmaEbert: do you have dircache and/or automatic update of the database enabled?
01:12:56Eberti may, let me check
01:14:34Ebertauto update of the database was on, i turned it off
01:14:55Eberti cant find dircache, do you mean if it updates to Ram or not?
01:15:52 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier.
01:16:53pixelmano, it's not a database setting, it's a general file cache. Should be in systems and disk settings
01:17:42Ebertgot it. directory cache was on. should i turn it off?
01:19:29pixelmathere were reports of the two settings causing disk thrashing at boot but I thought it was fixed. You could try. Do you have a lot of files (maybe small ones) or did you check your disk for fragmentation?
01:21:13pixelmaturning dircache off means that you will have to wait for a disk spinup when browsing if it isn't already spinning, makes it feel a bit unresponsive
01:21:37Ebert5000 files in dircache
01:22:53Eberti reinitialized the database just in case. have to wait for that now. I dont see how my disk could be fragmented
01:23:06Ebertthere are 3gbs left empty on it
01:24:35pixelmaand without dircache database autoupdate is slow IIRC (I don't use the database so wouldn't know and I also don't use dircache since my player is flash based so quick enough for me)
01:26:25pixelmajust from what I read here and there :)
01:27:17Ebertthanks for your info :) if turning the database off is what it takes, im ok with that
01:29:18Ebertautoupdate may happen more often and be causing it to thrash in the newer versions
01:31:46pixelmaI thought it should only happen if the content on your player had changed
01:33:10Ebertyea that would seem logical, but it seemed to be updating constantly
01:33:37EbertI fixed it now after redoing the database and turning auto-update off
01:33:46Ebertwhew, now I have battery life
01:36:10pixelmamaybe you could try asking about other people's experience, ones who really use it and at a different time of day
01:37:49Ebertok. sometimes its hard to get input here if you dont know the concise lingo. But you've given me hope heh
01:42:43 Quit DataGhost (Ping timeout: 240 seconds)
01:46:38 Quit krabador (Quit: Sto andando via)
01:50:18 Quit guysoft22 (Read error: Connection reset by peer)
01:50:32 Join guysoft22 [0] (
01:51:30***Saving seen data "./dancer.seen"
02:04:38 Quit Rob2222 (Read error: Connection reset by peer)
02:04:49 Join Rob2222 [0] (
02:16:52 Join Rob2223 [0] (
02:20:50 Quit Rob2222 (Ping timeout: 276 seconds)
02:26:37 Quit moos (Ping timeout: 252 seconds)
02:37:54 Quit einhirn (Ping timeout: 245 seconds)
02:48:02 Join stormdragon2976 [0] (~stormdrag@
02:51:58stormdragon2976Would it be difficult to write a bash script to generate all the phrases, menus, letters, etc for the rockbox menus. Are they the same format as .talk files?'
02:56:14saratogaprobably not that difficult, the VBS script to do it was short enough
02:57:19stormdragon2976Do I just need to create a series of .talk files and put them in the right directory?
02:57:55saratogai think the wiki said you need to put each voice file in the directory it should be read in, but its probably worth checking that yourself
02:59:11saratogaits probably easiest to just install a screen reader that works with QT and then use our tools to do it though
03:01:10stormdragon2976no Linux screen reader works with QT yet. It may be a while lol. So I am trying to just build Rockbox tools in to Nautilus. It's going very well so far.
03:01:16 Quit M3DLG (Ping timeout: 252 seconds)
03:01:55stormdragon2976Windows isn't really an option.
03:13:49 Quit MethoS- (Remote host closed the connection)
03:16:42 Join ischeria1 [0] (
03:17:38 Part stormdragon2976 ("Leaving.")
03:19:39 Quit ischeriad (Ping timeout: 260 seconds)
03:33:21 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
03:36:55 Join S_a_i_n_t_ [0] (S_a_i_n_t@
03:37:06 Quit S_a_i_n_t (Ping timeout: 264 seconds)
03:37:16 Nick S_a_i_n_t_ is now known as S_a_i_n_t (S_a_i_n_t@
03:51:33***Saving seen data "./dancer.seen"
04:04:29 Nick fxb is now known as fxb__ (
04:09:42 Part Ebert
04:10:01 Quit TheSeven (Ping timeout: 240 seconds)
04:13:51 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:14:06 Nick JdGordon1 is now known as JdGordon (
04:17:28kisakin the game bubbles, if you hit select before the round has begun, then the first marble is launched immediately once the round starts
04:18:06 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:18:06kisakit would be nice if select dismissed the message instead
04:18:06 Quit pixelma (Disconnected by services)
04:18:10 Quit amiconn (Disconnected by services)
04:18:12 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:18:25 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:18:32 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:19:26 Quit arbingordon (Read error: Connection reset by peer)
04:28:36 Quit adnyxo (Ping timeout: 246 seconds)
04:29:23CIA-5New commit by jdgordon (r26095): the fm presets list should wrap.. so make that work
04:42:54 Join FlynDice [0] (~FlynDice@
04:49:46 Join dys`` [0] (
04:50:27 Join Barahir [0] (
04:51:43 Quit Barahir_ (Read error: Operation timed out)
04:52:14 Quit dys` (Ping timeout: 276 seconds)
04:52:29 Join ArcticFang [0] (
04:53:08 Quit ArcticFang (Client Quit)
05:27:20 Join DerPapst [0] (
05:34:01 Quit DerPapst (Quit: Leaving.)
05:36:30 Join anewuser [0] (anewuser@unaffiliated/anewuser)
05:38:58 Quit Horscht (Quit: Verlassend)
05:40:17 Join antil33t [0] (
05:51:34***Saving seen data "./dancer.seen"
05:55:58 Quit Strife89 (Quit: Bedtime! ZZZZZZzzzzzzzzzzz.........)
05:59:46 Join newbie4rockbox [0] (
06:01:03 Part Boldfilter
06:02:24 Quit newbie4rockbox (Client Quit)
06:02:31 Join webguest79 [0] (
06:05:09 Quit webguest79 (Client Quit)
06:06:11 Join pleezwork [0] (
06:07:06pleezworkhello world
06:08:09CIA-5New commit by jdgordon (r26096): count is being used as the value of the last item to show, so rename it to not be confusing
06:11:30pleezworkCIA-5 are you a person?
06:14:54 Quit FlynDice (Remote host closed the connection)
06:17:47 Join FlynDice [0] (~FlynDice@
06:17:48 Quit pleezwork (Quit: CGI:IRC (EOF))
06:24:23 Quit FlynDice (Quit: Gotta go fly!)
06:24:45 Join FlynDice [0] (~FlynDice@
06:29:26 Quit anewuser (Quit: for SELL 2 by the price of 1 now!)
06:47:16 Quit JdGordon (Changing host)
06:47:16 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
06:50:24 Quit umrain (Ping timeout: 240 seconds)
07:29:57 Quit ssorgatem (Read error: Network is unreachable)
07:34:21 Join ssorgatem [0] (
07:42:47 Quit shai (Quit: Leaving)
07:47:18 Quit guysoft22 (Read error: Connection reset by peer)
07:47:29 Join kifo [0] (
07:47:42 Quit kifo (Client Quit)
07:48:12 Join kifotv [0] (
07:48:15 Join GuySoft [0] (
07:48:33kifotvtest test
07:49:45 Quit kifotv (Client Quit)
07:51:25 Quit GuySoft (Read error: Connection reset by peer)
07:51:38***Saving seen data "./dancer.seen"
07:51:40 Join GuySoft [0] (
07:52:44 Quit mt (Ping timeout: 240 seconds)
07:53:20 Join shai [0] (
07:53:20 Join kifo [0] (
07:54:46kifoHi, I am trying to fix a Sansa Fuze V2 with the current rev to have the wheel light turned completely off. Even with the button press wheel light time out set to off, it still seems to be controlled by disk activity or something, and I'm not sure how to fix that.
07:55:12 Join stoffel [0] (
07:55:31kifoIs there anything else that the wheel light is controlled by? It's the flickering wheel light that's really annoying, but I can't seem to find any way to disable it.
07:56:00 Join mt [0] (~mtee@rockbox/developer/mt)
07:58:28tmztthat's a debugging feature I think
07:58:48tmztbut it might be hardcoded
07:58:53tmztand not in the debug menu
08:00:43kifoI just cleared the option in the config file, then reset it with the current rev, and so far it looks fixed. I'll have to check the debug menu.
08:00:55 Quit GuySoft (Read error: Connection reset by peer)
08:01:05 Join GuySoft [0] (
08:01:22 Quit GuySoft (Read error: Connection reset by peer)
08:01:36 Join GuySoft [0] (
08:03:21kifoIt's not in the debug menu, but it looks like it's fixed by just clearing the option and resetting it.
08:03:32 Quit kifo (Quit: leaving)
08:05:25 Join mikroflops_ [0] (
08:09:02 Quit mikroflops (Ping timeout: 240 seconds)
08:24:16 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:30:31 Join ender` [0] (
08:36:25 Quit TheSeven (Ping timeout: 248 seconds)
08:38:54 Join flydutch [0] (
08:45:44 Join Forsaken_Boy [0] (~chatzilla@
08:54:23 Join kugel [0] (~kugel@rockbox/developer/kugel)
08:57:55 Quit stoffel (Remote host closed the connection)
08:59:01kugelgevaerts: I have a working solution for the mingw sims
08:59:30ZagorS_a_i_n_t: is all of that rebuilt every time??
09:00:09kugelthe problem lies in PollEvent. it implicitely calls SDL_PumpEvent, which you are only allowed to call from the thread that calls SDL_Init(SDL_INIT_VIDEO)
09:00:31S_a_i_n_tZagor: Yep
09:00:52ZagorS_a_i_n_t: can you just type "make" and pastebin the output?
09:01:12Zagoryour pastebin actually has changed files in it
09:01:14pixelmaI haven't built without changes in a while
09:01:15 Join petur [0] (~petur@rockbox/developer/petur)
09:02:08*kugel wonders what problem is being talked about
09:02:23S_a_i_n_tMy horrible CygWin
09:03:06pixelmaI remember it saying "nothing to be done" though
09:03:33S_a_i_n_tI've seen "Nothing to be done for 'All' " once, ever
09:03:36pixelmacan't tell if anything has changed recently and can't test now from work
09:04:54S_a_i_n_tI just ran configure/make now, and it's pumping out a huge list again...I suspect this may take some time, but the important thing is that in between this make, and the last one I posted I haven't updated *any* source files.
09:05:04S_a_i_n_tso, in theory, nothing should need to be done.
09:05:36 Join DerPapst [0] (
09:06:45S_a_i_n_tI updated to current rev just before I ran that last make/make fullzip I pastebinned (you can see the files that were updated up the top), it looks to me like it compiled a hell of a lot more than just the files that were touched by the update.
09:07:23ZagorS_a_i_n_t: why are you running configure? just run make *that* my problem?
09:07:50Zagormost likely
09:08:09Zagoryou should only run congfigure when you change the target
09:08:17S_a_i_n_tHmmm...ok, I'll give that a whirl when this finishes.
09:08:20Zagor(or when make tells you to)
09:11:49pixelmaS_a_i_n_t: you also need "make fullzip" only when your fonts changed", otherwise "make zip" or "make 7zip" will do
09:12:06pixelmathat was one too many "
09:12:25 Quit bieber (Ping timeout: 260 seconds)
09:12:39 Join bieber [0] (
09:12:48 Quit Forsaken_Boy (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
09:15:32kugelS_a_i_n_t: in most situations you want a complete recompile if the Makefile changes, so just stop running configure :)
09:18:49 Quit phanboy4 (Ping timeout: 245 seconds)
09:19:01pixelmaa complete recompile needs a configure again if there were makefile changes, or am I missing something?
09:20:26 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
09:21:58Zagorconfigure _creates_ the Makefile, so quite obviously it becomes modified
09:23:23 Join B4gder [0] (~daniel@rockbox/developer/bagder)
09:25:42pixelmaI meant changes that needs a new makefile
09:28:52Zagorthere is a check if tools/configure has been changed since your Makefile was generated. if so, you get a message "tools/configure is newer than Makefile. You should run 'make reconf'"
09:29:33 Quit kugel (Ping timeout: 246 seconds)
09:31:28*S_a_i_n_t facepalms...yep, that was it.
09:31:43S_a_i_n_tmental note: Don't run configure every time...
09:32:01pixelmaZagor: I know, I was just wondering about kugel's statement
09:34:20 Join n1s [0] (~n1s@rockbox/developer/n1s)
09:34:41 Quit [CGL] (Quit: ---- \m/ d(>.<)b \m/)
09:43:05 Join lpereira [0] (
09:43:34 Quit GuySoft (Read error: Connection reset by peer)
09:43:38 Join GuySoft [0] (
09:48:11 Quit DerPapst (Quit: Leaving.)
09:48:26 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
09:51:39***Saving seen data "./dancer.seen"
09:58:09CIA-5New commit by jethead71 (r26097): I think I intended a different expression.
10:07:22 Join hebz0rl [0] (
10:11:14 Join mitk [0] (~mitk@
10:12:09CIA-5New commit by nls (r26098): Update the notes for which compression levels of ape are realtime and fix the location of rockbox version in the bug reporting instructions
10:13:13 Quit flydutch (Quit: /* empty */)
10:13:19n1spixelma, bluebrother; any idea why the appendix begins with a blank page?
10:13:41n1salso, does the NODO list really belong in the manual?
10:13:54 Join DerPapst [0] (
10:17:26 Join flydutch [0] (
10:21:54JdGordonn1s: it really really doesnt!
10:22:46gevaertsgsoc students: you're on schedule with the google paperwork?
10:23:16 Quit JdGordon (Quit: Leaving.)
10:24:23pixelman1s: why not? (re. the NoDo list, I think it would be good if users know it too if they ask for features) - and is the blank page in every manual?
10:26:26n1spixelma: i dont' know, it feels out of place to be but i don't feel strongly about it, the blank page is in every pdf i checked
10:29:10n1salso the note about usb host being impossible isn't entirely accurate
10:29:35S_a_i_n_tWhat about just making the NODO list a sticky topic in Feature Ideas on the Forum?
10:29:56S_a_i_n_tI seem to feel that the manual isn;t really the right place for it myself...
10:30:25 Join Jaykay [0] (
10:31:05*gevaerts points out that it's a wiki page in the first place, so he doens't see why it should be copied to the forums
10:31:44S_a_i_n_tShouldn;t that be the same for the Manuals then also?
10:32:03gevaertsI think so, yes
10:32:40 Join Parsi [0] (~Maysam@
10:32:50S_a_i_n_tI definitely think it should be available for users to see (and it is), I just don't feel that the Manual is the right place for it.
10:35:31pixelman1s: just checked two manuals (Mini 1G and H120) and the latter had the blank page, not the former. It's a bit weird indeed though
10:35:31n1syeah, i think a link to the wiki is nicer, we do this in some other places in the manual too
10:36:09n1spixelma: indeed
10:36:55n1spixelma: are you opposed to replacing the NODO list in the manual with a link to the wiki?
10:38:32 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
10:39:35S_a_i_n_tI was reading the in the forums about a feature idea re: pausing playback if ambient noise is too high (mic enabled targets) such as a passing vehicle etc. Hasn't this idea already been discussed and thrown out? IIUC most users put their DAPs in their pockets, and friction between the DAP and the fabric would almost certainly make enough noise to keep the player paused all the time.
10:40:31n1sS_a_i_n_t: it comes up every now and then and the idea has a number of *big* flaws that the suggesters tend to ignore even when they are pointed out to them
10:41:05S_a_i_n_tn1s: I thought I had seen it least I'm not insane. ;)
10:41:25n1swell, usually they suggest active noise cancellation but
10:42:03S_a_i_n_tIt's a nice idea, but I very sincerely doubt it would work as intended. Something like coughing for instance would probably pause the player also.
10:42:03n1sseveral players can not record and play back at the same time though so neither this nor noise cancellation will work on them
10:42:35S_a_i_n_tyes, the idea is nice though...dreams are free :D
10:42:48 Join funman [0] (~fun@rockbox/developer/funman)
10:42:51n1sand for noise cancellation you need the mic very close to the headphone and very low latency in the sound processing
10:43:51 Join max242 [0] (
10:44:16funmanjhMikeS: <- can you test this on a discharged e200v1 and see if it charges ? (just monitor battery debug menu 3rd screen)
10:45:31funmanit seems endofch irq monitoring doesn't need to be disabled/reenabled, but can just be left enabled.
10:46:14 Quit GuySoft (Read error: Connection reset by peer)
10:46:14 Join guysoft22 [0] (
10:46:20funmanthe interrupt needs to be cleared when enabling the charger though, I noticed it will also happen on charger unplug, regardless if the battery is charged or not
10:47:02 Quit guysoft22 (Read error: Connection reset by peer)
10:47:10 Join guysoft22 [0] (
10:47:14 Part Parsi
10:47:28 Join JdGordon [0] (~jd@rockbox/developer/JdGordon)
10:48:07funmanmoving the read of charger status (plugged/unplugged) is useful for the as3525 part of my diff (not pasted)
10:50:50funmanjhMikeS: please check if it still charges after unplugging the charger and replugging also
10:52:11funmanhmmm the as3525 datasheet says 'The interrupt must not be enabled if the charger block is'
10:53:00funmanAfter turning off the charger, IRQ will be released
10:53:08funman^−− this is as3514 datasheet
10:53:22funmanI understand 'released == acknowledged' here
10:57:31 Join einhirn [0] (
10:59:00 Join efyx [0] (
10:59:28 Join LinusN [0] (~linus@rockbox/developer/LinusN)
11:01:33 Join mikroflops [0] (
11:01:54funmanthe as352x source code for as3515/as3517 leaves the interrupt enabled regardless if the charger is disabled or not, but does 'clear the interrupt flag caused by the enable of the charger'
11:02:05funmanso it looks right to me
11:02:31funmanjust needs confirmation that it works on as3514/PP
11:05:34 Quit mikroflops_ (Ping timeout: 265 seconds)
11:08:00S_a_i_n_t605KB for rockbox.ipod..... :/
11:17:33 Quit max242 (Quit: CGI:IRC (Ping timeout))
11:25:49 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:29:08 Join swilde [0] (
11:34:04pamauryfunman: does the code for the as3525 still generates thousands of audio interrupts or not ? (read: should I try to detect usb connection with it or not ?) :)
11:34:20funmanas3525v2, last time i tried yes it did
11:35:12funmani think for now you should put a timer like if(current_tick > HZ*10) return USB_INSERTED; else return USB_EXTRACTED;
11:37:29pamaurywhat a lovely hack :)
11:37:44funmanI don't get what's the build condition for rockboy
11:38:05funmanif it builds & runs on the 112x64 monochrome archos recorder, it will build & run everywhere?
11:40:06 Join matsl [0] (~matsl@
11:45:30funman <- screenshots of pokemon, <- screenshot of tetris
11:46:43pamaurythe clip+ is not great for gaves because of the color chang
11:46:51n1sfunman: awesome!
11:47:51 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
11:48:18 Quit dfkt (Disconnected by services)
11:48:21 Nick dfkt_ is now known as dfkt (dfkt@unaffiliated/dfkt)
11:48:27funmanit's realtime, sound is perfect
11:51:18funman <- pokemon screendump, hardly readable but if you know the game it's ok
11:51:43***Saving seen data "./dancer.seen"
11:52:02pamauryhaha, that's completely unreadable
11:55:42funmanhm why can't i see rockboy in the online archos recorder manual ? it should be there
11:57:15funmani guess rockboy would run on clipv1 too but with a bigger plugin buffer
11:58:01n1shmm, test_codec seems to be buggy if i try to make several dir logs in one run, for example a "speed test folder" and "checksum folder" either only writes one log or oone log and one empty file...
12:01:24 Quit guysoft22 (Read error: Connection reset by peer)
12:01:39 Join guysoft22 [0] (
12:01:51CIA-5New commit by funman (r26099): rockboy: add a Clip keymap ...
12:01:56 Quit TheSeven (Ping timeout: 260 seconds)
12:03:17S_a_i_n_tJdGordon: Anyone: If you're considering commiting FS #11270 I have a suggestion for it, it'd be *really* nice if it automagically went to the currently playing album in the pictureflow if the hotkey is pressed in the WPS when playback has been started by the database/filebrowser.
12:03:37S_a_i_n_tAt the moment, it only remembers the current album if playback is started by the plugin.
12:03:46S_a_i_n_tOther than that, it's really quite cool.
12:04:24 Quit guysoft22 (Read error: Connection reset by peer)
12:04:39 Join guysoft22 [0] (
12:05:41 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
12:06:06funmann1s: how slow is rockboy on archos recorder?
12:06:18pixelmafunman: I guess it could be an overlay plugin then as it is on the Recorders (use the audio buffer)
12:06:26pixelmafunman: very
12:07:13funmanif speed isnt' a problem i'll just enable rockboy everywhere (except for archosplayer i guess)
12:07:52pixelmabut n1s's post doesn't take the fact into account that the Recorders seem to have quite some more amount of usable RAM available (as the firmware doesn't take us much as we found out when speaking of voice)
12:08:12pixelmaon the Ondios you also have a buttonmap problem (not enough buttons)
12:08:37pixelmaand the greylib looks way nicer :)
12:08:50funmanrockboy works with greylib?
12:09:35pixelmahmm.. not sure, I thought so but may be wrong
12:12:01funmanmodels for which rockboy is disabled are: ipod1g2g, 3g, 4g, mini, iaudiom3, zenvision, mpio, mrobe100, yh920, clips and m200v4
12:12:30funmannot counting archos player and ondiosp because we need LCD_BITMAP & enough buttons
12:13:49 Join watto [0] (~watto@
12:14:01pixelmathe old Ipods have the same problem re. buttons (on newer the touch sensitive wheel is used in "zones" to get more "buttons")
12:14:25funmanall the ones I named don't have enough buttons to run rockboy ?
12:14:34funmanall the ipods*
12:14:43pixelmaand now I seem to remember that the greyscale and mono displays need specific LCD adaptations
12:16:46 Quit pamaury (Ping timeout: 264 seconds)
12:17:40 Join JohannesSM64 [0] (
12:18:04 Join kugel [0] (~kugel@rockbox/developer/kugel)
12:19:52funmannot enough memory on clipv1
12:20:54funmanthe condition i guess is: LCD_BITMAP && (MEM > 2 || HWCODEC) && (not in list of models which lack buttons)
12:22:05funmanI guess the 'MEM <= 8' is not 'MEM < 8' to support the 8MB archos mods ?
12:26:21linuxstbfunman: The 4g ipod's wheel is the same as later ipods. The problem there is (I think) that no-one has adapted the Rockboy lcd code for that ipod's greyscale framebuffer format.
12:26:46linuxstbThe 2nd gen mini may be the same reason.
12:26:56 Join GeekShado_ [0] (
12:28:29funman1st gen mini use the same pixel format than the 2nd gen
12:29:51n1spixelma: no i forgot about that, but the clips have ~300k of iram too
12:30:16 Quit GeekShadow (Ping timeout: 260 seconds)
12:30:34funmanthe codecs go there
12:31:06n1sfunman: can't plugins use that at all?
12:31:26n1sah, is it any faster than dram?
12:31:34funmanwe could perhaps use some dirty tricks when playback is disabled
12:31:43funmani'm not sure
12:32:18n1son the other iram targets, the plugins and codecs share the same iram section so codecs cannot run when a plugin uses iram
12:32:23 Quit GeekShado_ (Ping timeout: 240 seconds)
12:32:51 Join M3DLG [0] (~M3DLG@
12:32:55funmanon clips & as3525v2 we trick rockbox to think there's no iram available for codecs & plugins
12:33:08funmaninstead we declare the whole codec buffer at the location of iram
12:33:27 Join bongo [0] (
12:33:35funmanso no copying of iram sections is involved: those (iram sections) do not exist for codecs since everything is in iram
12:34:10n1smaybe you could only trick the codecs and have the plugins work as on other targets
12:34:39funmanwe don't need to use iram with plugins
12:34:52n1sbtw, midiplaer will not work for any reasonable music with this little ram either
12:35:02funmanfor the few plugins which might benefit from speed, the fast CPU is enough
12:35:19n1sfunman: oh, ok, i though pixelma said you had too little ram but i might have misread
12:36:15funmanyes you misread, because *i* said that :p
12:36:53funmani think both the plugin buffer & the audiobuffer are too small to hold rockboy on clipv1, but the combination of the 2 might be enough
12:36:54n1sso you do have too little ram?
12:37:12n1sah, yes
12:38:18funmanoh but it links on c200v2 ? o:
12:38:36funmanthen i think just enlarging the plugin buffer a bit might do the trick
12:38:55funmannow i just need to know if rockboy will run on m3, piod, yh920, and mrobe100
12:39:39jhMikeSfunman: everything was find on e200v1. If you're reading the int status in an actually ISR, then you're probably stealing the EOCH indicator from powermgmt
12:40:04funmanjhMikeS: 'everything was fine' -> you tested the patch?
12:40:06jhMikeSs/actually/actual >:[
12:40:23bongofunman: M3 have no main LCD just the remote IIRC
12:40:36jhMikeSfunman: oh, no...I looked back and only saw that in the history buffer here
12:40:41n1sbongo: yup
12:41:09bongothen no rockboy :)
12:41:26jhMikeSfunman: no, it's there...just way, way back :)
12:41:47funmanjhMikeS: btw reading CHG_STATUS will steal the indicator, i know of this problem but i'll correct it after (one step at a time)
12:41:59funmanbut other plugins work on the remote?
12:42:04funman(re: m3)
12:42:24 Quit guysoft22 (Read error: Connection reset by peer)
12:42:37linuxstbbongo: IIUC, as far as Rockbox is concerned, the M3's remote LCD _is_ the main LCD.
12:42:40 Join guysoft22 [0] (
12:42:52*jhMikeS thinks perhaps, just take all the steps since the behavior is well known (to me anyway)
12:42:52n1sbongo: why not?
12:43:25bongolinuxstb: the remote LCD is quite small no?
12:43:26funmanjhMikeS: so my 1st patch looks ok to you ?
12:43:42funmanbongo: bigger than clip screen at least ;)
12:43:43linuxstbbongo: Bigger than some target's main LCD I think.
12:44:00bongo... with working rockboy?
12:44:09 Quit JdGordon (Quit: Bye)
12:44:24jhMikeSfunman: you disabled the interrupt enable? I don't think it'll work actually.
12:44:30funmanbongo: for some value of working
12:44:35funmanjhMikeS: no i enable it once for all at init
12:44:42n1sbongo: yeah, it's bigger then the clip's main lcd
12:44:51funmanthe as352x linux code does just that
12:45:24bongocrazy guys :)
12:45:28 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
12:45:39funmanand i still get the interrupt (the false one anyway, my 2 fuzes are charging and i'll soon see if it happens at real end of charging)
12:45:47 Quit M3DLG (Ping timeout: 260 seconds)
12:46:50jhMikeSfunman: ok, missed that
12:47:22funmani just want to get the PP stuff confirmed because then I'm free to experiment on as3525 code
12:49:54jhMikeSpatch hates the git patch format
12:50:21n1sB4gder: have you noticed that the frontpage says "Page was last modified __FILE_DATE__ The Rockbox Crew" on the bottom?
12:50:46B4gderaha, nice ;-)
12:50:47 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
12:51:48funmanjhMikeS: patch -p1 ?
12:52:00funmanjhMikeS: sometime pastebins eat the last newline
12:53:40 Join wodz [0] (
12:58:28 Quit funman (*.net *.split)
12:58:28 Quit n1s (*.net *.split)
12:58:29 Quit Zarggg (*.net *.split)
12:58:29 Quit markun (*.net *.split)
12:58:29 Quit AlexP (*.net *.split)
12:58:30 Quit Guest951 (*.net *.split)
12:58:30 Quit rasher (*.net *.split)
12:58:30 Quit pamaury (*.net *.split)
12:58:30 Quit lpereira (*.net *.split)
12:58:30 Quit mc2739 (*.net *.split)
12:58:30 Quit jae (*.net *.split)
12:58:30 Quit dfkt (*.net *.split)
12:58:30 Quit evilnick (*.net *.split)
12:58:30 Quit avacore (*.net *.split)
12:58:30 Quit aevin (*.net *.split)
12:58:30 Quit gevaerts (*.net *.split)
12:58:30 Quit JdGordon (*.net *.split)
12:58:30 Quit petur (*.net *.split)
12:58:30 Quit ssorgatem (*.net *.split)
12:58:30 Quit jordan` (*.net *.split)
12:58:30 Quit n17ikh (*.net *.split)
12:58:30 Quit linuxguy3 (*.net *.split)
12:58:31 Quit Farthen (*.net *.split)
12:58:31 Quit stavrob (*.net *.split)
12:58:31 Quit dionoea (*.net *.split)
12:58:31 Quit Torne (*.net *.split)
12:58:31 Quit DerPapst (*.net *.split)
12:58:31 Quit mitk (*.net *.split)
12:58:31 Quit FlynDice (*.net *.split)
12:58:31 Quit Rob2223 (*.net *.split)
12:58:31 Quit shaggy-h (*.net *.split)
12:58:31 Quit detaos (*.net *.split)
12:58:31 Quit togetic (*.net *.split)
12:58:31 Quit Bagder (*.net *.split)
12:58:31 Quit lostlogic (*.net *.split)
12:58:31 Quit jhMikeS (*.net *.split)
12:58:31 Quit ischeria1 (*.net *.split)
12:58:31 Quit sevard (*.net *.split)
12:58:32 Quit rhodan (*.net *.split)
12:58:32 Quit Topy44 (*.net *.split)
12:58:32 Quit soap (*.net *.split)
12:58:32 Quit LinusN (*.net *.split)
12:58:32 Quit ps-auxw (*.net *.split)
12:58:32 Quit ranma (*.net *.split)
12:58:32 Quit scorche|sh (*.net *.split)
12:58:32 Quit FOAD (*.net *.split)
12:58:32 Quit tmzt (*.net *.split)
12:58:32 Quit BlakeJohnson86 (*.net *.split)
12:58:32 Quit Hadaka (*.net *.split)
12:58:32 Quit daurnimator (*.net *.split)
12:58:32 Quit ThomasAH (*.net *.split)
12:58:32 Quit fxb__ (*.net *.split)
12:58:32 Quit watto (*.net *.split)
12:58:32 Quit hebz0rl (*.net *.split)
12:58:32 Quit bieber (*.net *.split)
12:58:32 Quit dys`` (*.net *.split)
12:58:32 Quit MagusG (*.net *.split)
12:58:32 Quit advcomp2019_ (*.net *.split)
12:58:33 Quit sharp (*.net *.split)
12:58:33 Quit blithe (*.net *.split)
12:58:33 Quit CIA-5 (*.net *.split)
12:58:33 Quit maraz (*.net *.split)
12:58:33 Quit swilde (*.net *.split)
12:58:33 Quit Jaykay (*.net *.split)
12:58:33 Quit mt (*.net *.split)
12:58:33 Quit amiconn (*.net *.split)
12:58:33 Quit pixelma (*.net *.split)
12:58:33 Quit solexx (*.net *.split)
12:58:33 Quit Zagor (*.net *.split)
12:58:33 Quit topik (*.net *.split)
12:58:33 Quit matsl (*.net *.split)
12:58:34 Quit Barahir (*.net *.split)
12:58:34 Quit Galois (*.net *.split)
12:58:34 Quit kisak (*.net *.split)
12:58:34 Quit JohannesSM64 (*.net *.split)
12:58:34 Quit efyx (*.net *.split)
12:58:34 Quit einhirn (*.net *.split)
12:58:34 Quit polobricolo (*.net *.split)
12:58:34 Quit rvvs89 (*.net *.split)
12:58:34 Quit whydoubt (*.net *.split)
12:58:34 Quit guysoft22 (*.net *.split)
12:58:34 Quit Kitar|st (*.net *.split)
12:58:34 Quit wincent (*.net *.split)
12:58:34 Quit bluebrother (*.net *.split)
12:58:34 Quit Beta2K (*.net *.split)
12:58:34 Quit powell14ski_ (*.net *.split)
12:58:35 Quit slck (*.net *.split)
12:58:35 Quit pjm0616 (*.net *.split)
12:58:35 Quit sinthetek (*.net *.split)
12:58:35 Quit kugel (*.net *.split)
12:58:35 Quit bluefoxx (*.net *.split)
12:58:35 Quit krazykit (*.net *.split)
12:58:35 Quit tchan (*.net *.split)
12:58:35 Quit Unhelpful (*.net *.split)
12:58:35 Quit bzed (*.net *.split)
12:58:35 Quit simabeis (*.net *.split)
12:58:35 Quit Zambezi (*.net *.split)
12:58:35 Quit B4gder (*.net *.split)
12:58:35 Quit kenguest (*.net *.split)
12:58:35 Quit jfc (*.net *.split)
12:58:36 Quit YPSY (*.net *.split)
12:58:36 Quit wombat23 (*.net *.split)
12:58:36 Quit ved (*.net *.split)
12:58:36 Quit TheSeven (*.net *.split)
12:58:36 Quit ender` (*.net *.split)
12:58:36 Quit S_a_i_n_t (*.net *.split)
12:58:36 Quit RankFrank (*.net *.split)
12:58:36 Quit grndslm (*.net *.split)
12:58:36 Quit crwl (*.net *.split)
12:58:37 Quit ChanServ (*.net *.split)
12:58:37 Quit shai (*.net *.split)
12:58:37 Quit antil33t (*.net *.split)
12:58:37 Quit yosafbridge (*.net *.split)
12:58:37 Quit yawny (*.net *.split)
12:58:37 Quit Llorean (*.net *.split)
12:58:37 Quit xavieran (*.net *.split)
12:58:37 Quit Tuplis (*.net *.split)
12:58:37 Quit scorche (*.net *.split)
12:58:37 Quit Necos (*.net *.split)
12:58:37 Quit Kohlrabi (*.net *.split)
12:58:37 Quit saratoga (*.net *.split)
12:58:37 Quit mikroflops (*.net *.split)
12:58:37 Quit flydutch (*.net *.split)
12:58:37 Quit BeFalou (*.net *.split)
12:58:38 Quit Battousai (*.net *.split)
12:58:38 Quit nima (*.net *.split)
12:58:38 Quit MuscleNerd (*.net *.split)
12:58:38 Quit Guest20471 (*.net *.split)
12:58:38 Quit Kamyk_ (*.net *.split)
12:58:38 Quit wodz (*.net *.split)
12:58:38 Quit bongo (*.net *.split)
12:58:38 Quit GodEater (*.net *.split)
12:58:38 Quit parafin (*.net *.split)
12:58:38 Quit linuxstb (*.net *.split)
12:58:38 Quit kadoban (*.net *.split)
12:58:38 Quit bug2000 (*.net *.split)
12:58:38 Quit cYmen (*.net *.split)
12:58:38 Quit Utchybann (*.net *.split)
13:00:10 Join wodz [0] (
13:00:10 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
13:00:10 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
13:00:10 Join guysoft22 [0] (
13:00:10 Join bongo [0] (
13:00:10 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:00:10 Join JohannesSM64 [0] (
13:00:10 Join watto [0] (~watto@
13:00:10 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
13:00:10 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:00:10 Join matsl [0] (~matsl@
13:00:10 Join swilde [0] (
13:00:10 Join mikroflops [0] (
13:00:10 Join LinusN [0] (~linus@rockbox/developer/LinusN)
13:00:10 Join efyx [0] (
13:00:10 Join einhirn [0] (
13:00:10 Join funman [0] (~fun@rockbox/developer/funman)
13:00:10 Join Jaykay [0] (
13:00:10 Join flydutch [0] (
13:00:10 Join DerPapst [0] (
13:00:10 Join mitk [0] (~mitk@
13:00:10 Join hebz0rl [0] (
13:00:10 Join lpereira [0] (
13:00:10 Join n1s [0] (~n1s@rockbox/developer/n1s)
13:00:10 Join B4gder [0] (~daniel@rockbox/developer/bagder)
13:00:10 Join bieber [0] (
13:00:10 Join petur [0] (~petur@rockbox/developer/petur)
13:00:10 Join ender` [0] (
13:00:10 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
13:00:10 Join mt [0] (~mtee@rockbox/developer/mt)
13:00:10 Join shai [0] (
13:00:10 Join ssorgatem [0] (
13:00:10 Join FlynDice [0] (~FlynDice@
13:00:10 Join antil33t [0] (
13:00:10 Join Barahir [0] (
13:00:10 Join dys`` [0] (
13:00:10 Join amiconn [0] (quassel@rockbox/developer/amiconn)
13:00:10 Join pixelma [0] (quassel@rockbox/staff/pixelma)
13:00:10 Join S_a_i_n_t [0] (S_a_i_n_t@
13:00:10 Join ischeria1 [0] (
13:00:10 Join Rob2223 [0] (
13:00:10 Join solexx [0] (
13:00:10 Join Kitar|st [0] (
13:00:10 Join shaggy-h [0] (
13:00:10 Join wincent [0] (~wincent@rockbox/developer/wincent)
13:00:10 Join BeFalou [0] (~mamutoi@unaffiliated/befalou)
13:00:10 Join yosafbridge [0] (
13:00:10 Join detaos [0] (
13:00:10 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
13:00:10 Join yawny [0] (
13:00:10 Join togetic [0] (~togetic@unaffiliated/ibuffy)
13:00:10 Join ranma [0] (
13:00:10 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
13:00:10 Join Galois [0] (
13:00:10 Join bluefoxx [0] (
13:00:10 Join Beta2K [0] (
13:00:10 Join Zarggg [0] (
13:00:10 Join Bagder [0] (~daniel@rockbox/developer/bagder)
13:00:10 Join saratoga [0] (~9803c6dd@gateway/web/freenode/x-hlzjjzgdsdosrljp)
13:00:10 Join jordan` [0] (
13:00:10 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
13:00:10 Join krazykit [0] (~kkit@
13:00:10 Join RankFrank [0] (
13:00:10 Join evilnick [0] (
13:00:10 Join kenguest [0] (
13:00:10 Join rasher [0] (~rasher@rockbox/developer/rasher)
13:00:10 Join powell14ski_ [0] (
13:00:10 Join kisak [0] (
13:00:10 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
13:00:10 Join polobricolo [0] (
13:00:10 Join sevard [0] (sev@
13:00:10 Join rhodan [0] (~Quassel@2001:1608:12:2::38)
13:00:10 Join jfc [0] (
13:00:10 Join xavieran [0] (
13:00:10 Join parafin [0] (
13:00:10 Join rvvs89 [0] (~ivo@pdpc/supporter/base/rvvs89)
13:00:10 Join lostlogic [0] (~lostlogic@rockbox/developer/lostlogic)
13:00:10 Join whydoubt [0] (
13:00:10 Join YPSY [0] (
13:00:10 Join Topy44 [0] (
13:00:10 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
13:00:10 Join soap [0] (~soap@rockbox/staff/soap)
13:00:10 Join MagusG [0] (
13:00:10 Join Tuplis [0] (
13:00:10 Join Torne [0] (torne@rockbox/developer/Torne)
13:00:10 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
13:00:10 Join slck [0] (Venci@Slackware.SlackPix.Com)
13:00:10 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
13:00:10 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
13:00:10 Join Battousai [0] (~bryan@gentoo/developer/battousai)
13:00:10 Join grndslm [0] (
13:00:10 Join jae [0] (
13:00:10 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
13:00:10 Join bzed [0] (
13:00:10 Join wombat23 [0] (
13:00:10 Join markun [0] (~markun@rockbox/developer/markun)
13:00:10 Join cYmen [0] (
13:00:10 Join nima [0] (
13:00:10 Join n17ikh [0] (
13:00:10 Join scorche|sh [0] (~scorche@rockbox/administrator/scorche)
13:00:10 Join pjm0616 [0] (~user@
13:00:10 Join Guest951 [0] (
13:00:10 Join AlexP [0] (~ap@rockbox/staff/AlexP)
13:00:10 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
13:00:10 Join MuscleNerd [0] (
13:00:10 Join FOAD [0] (~dok@
13:00:10 Join bug2000 [0] (~bug@unaffiliated/bug2000)
13:00:10 Join linuxguy3 [0] (
13:00:10 Join kadoban [0] (
13:00:10 Join Farthen [0] (
13:00:10 Join avacore [0] (
13:00:10 Join simabeis [0] (
13:00:10 Join tmzt [0] (
13:00:10 Join aevin [0] (eivindsy@unaffiliated/aevin)
13:00:10 Join maraz [0] (
13:00:10 Join CIA-5 [0] (cia@
13:00:10 Join blithe [0] (~blithe@
13:00:10 Join sharp [0] (
13:00:10 Join scorche [0] (~scorche@rockbox/administrator/scorche)
13:00:10 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
13:00:10 Join stavrob [0] (
13:00:10 Join daurnimator [0] (daurnimato@freenode/staff/daurnimator)
13:00:10 Join Zambezi [0] (Zulu@unaffiliated/zambezi)
13:00:10 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
13:00:10 Join BlakeJohnson86 [0] (~bjohnson@2002:1876:a27b:0:227:13ff:fe65:1262)
13:00:10 Join Necos [0] (
13:00:10 Join Utchybann [0] (
13:00:10 Join Kamyk_ [0] (
13:00:10 Join crwl [0] (
13:00:10 Join ved [0] (
13:00:10 Join Hadaka [0] (
13:00:10 Join dionoea [0] (~dionoea@videolan/developer/dionoea)
13:00:10 Join Kohlrabi [0] (
13:00:10 Join Guest20471 [0] (
13:00:10 Join fxb__ [0] (
13:00:10 Join ThomasAH [0] (
13:00:10 Join ChanServ [0] (ChanServ@services.)
13:00:10 Join topik [0] (
13:00:10Mode"#rockbox +o ChanServ " by
13:00:10DBUGsent MODE #rockbox -o ChanServ
13:00:16S_a_i_n_tI have *NO* idea how it could be done, but I keep thinking about a conditional tag for font height in pixels. I figure theme authors could accomodate themes for multiple different layouts depending on the users preferred font that way.
13:00:16***Server message 484: 'logbot ChanServ #rockbox :Cannot kick or deop a network service'
13:00:22S_a_i_n_tJdGordon: So, apply that patch...if everything seems to work as it should, it's all good?
13:00:37JdGordonyeah, although I doubt it will just work :)
13:00:44JdGordonI dont remember the issues why I gave up last time
13:00:49amiconnfunman: You don't necessarily need a bigger plugin buffer. You can make use of the overlay mechanism
13:00:51JdGordonand no, that font height thing doesnt make sense...
13:01:05funmanamiconn: unless the audiobuffer is smaller than the plugin buffer
13:01:22S_a_i_n_tWell, it makes'd just be *really* hard to do I'd imagine.
13:01:26amiconnHmm, true. How big is the plugin buffer?
13:02:04JdGordonfunman: if that is the case, then wouldnt it make sense to make the plugin buffer TINY and always use the overlay?
13:02:10funmansmall enough for rockboy not to build, i think i just have to enlarge it a bit at the expense of audio buffer
13:02:34funmanJdGordon: unless you want to use plugins & codecs at the same time
13:02:38amiconnrockboy is disabled for the greyscale ipods and the M3 because noone wrote the necessary scaling routine for the pixel formats
13:03:00JdGordonS_a_i_n_t: patch doesnt apply, I'll ping you when its ready again :)
13:03:04amiconnI'd rather *lower* the plugin buffer on the clip
13:03:36S_a_i_n_tJdGordon: just about to give you a yell saying it needs syncing ;)
13:03:56funmanamiconn: thanks for the details, now that leaves mrobe100 and yh920 <- perhaps those just need a keymap
13:04:02amiconnbongo: The remote lcd is used as the main one for the M3. That's not the problem with rockboy
13:04:19n1sJdGordon: using the overlay stops playback
13:05:00funmanthe clip plugin buffer is already at a minimum
13:05:06JdGordonyes, but didnt someone say the clip's audio buffer is ~400KB only? surely halving the plugin buffer makes alot of sense
13:05:10amiconnMinimum being?
13:05:41amiconnImo the plugin buffer for the clip should be 48KB or 64KB
13:06:01funmanmimum at which all enabled plugins link
13:06:11JdGordonespecially with its screen being tiny means plugins wont be bloated much anyway
13:06:30amiconnThe larger plugins are usually those which would be acceptable as an overlay
13:07:01amiconnOn archos, the plugin buffer is 32KB. The 48KB suggestion is to account for arm code usually being larger than SH1 code
13:07:48funmanfft.rock 147K
13:08:20 Quit wincent (Ping timeout: 240 seconds)
13:08:24*linuxstb looks around for archivator...
13:09:09S_a_i_n_tJdGordon: This may not be something to think about right now, but re: fs#11133, could there be some tag to switch between "follow tag draw order" and "parse then draw as a lump" (like it does now)?
13:09:17n1shmm, fft uses O3, that's probalby part of the reason it's so big
13:09:23S_a_i_n_tI imagine it could make some themes look *really* weird ;D
13:09:24jhMikeSfunman: says it's far so good. it wasn't that empty so it shouldn't be long
13:09:29JdGordonS_a_i_n_t: no
13:09:30linuxstbn1s: Plus const.h (code in a header...)
13:09:58linuxstb(or rather, data in a header...)
13:10:07funmanjhMikeS: great, did you unplug/replug to be sure interrupt still happens after charging has been disabled/enabled again ?
13:10:35n1slinuxstb: it seems const.h is only included in one c file
13:11:27jhMikeSfunman: the batt is low enough and so it restarts
13:11:28n1sothoth some of those arrays don't need to use as wide types as they do
13:12:43jhMikeSfunman: now, what's important is that it stops properly...moreso than charging
13:13:02JdGordonS_a_i_n_t: ok, its resynced
13:13:44JdGordoncabbie on e200 seems correct...
13:14:21funmanfuzev1 is at 98% and voltage is flat but still charging, doesn't look good :o
13:15:02jhMikeSummm...constant voltage phase? too bad it can't read the current
13:15:09jhMikeSwhat's the voltage?
13:15:16linuxstbn1s: None of them need to be 32-bit, do they? The largest value I can see in the whole file is 32768.
13:15:40funmanbetween 4.184 & 4.189
13:15:55jhMikeSthat seems low actually
13:16:22jhMikeSI'd expect a lift to 4.210-4.214 or so
13:16:29n1slinuxstb: exactly
13:16:37n1syou fix, or I fix?
13:16:39funmanperhaps it's different between models?
13:17:10jhMikeSwouldn't expect it if the reg is set to 4.20 and the charge current is high enough
13:17:54linuxstbn1s: You can fix - I should be working at the moment...
13:18:04 Quit Jaykay (Ping timeout: 258 seconds)
13:18:31n1syeah, that saved 63k
13:19:00 Quit slck (Read error: Operation timed out)
13:19:14jhMikeSfunman: these currents seem sort of low (100mA)? for what capacity?
13:20:11wodzI am about to integrate byte swap into ata-as-coldfire.S. Could You review this patch? FS #11283 (amiconn especially)
13:22:05*S_a_i_n_t discovers that fft displays nothing in "linear scale - spectogram" :/
13:22:38S_a_i_n_t"bars" on laogarithmic used to be broken, now that works and spectogram doesn't. :S
13:22:46funmanjhMikeS: sorry i have to go
13:22:50S_a_i_n_t*Nano1g/current SVN
13:23:05 Quit matsl (Quit: Leaving)
13:23:25 Quit funman (Quit: free(random());)
13:26:41n1swodz: i'd suggest using a more soecific #define such as ATA_SWAP_WORDS or something and maybe turning those long byteswapping sequences into macros?
13:27:46wodzn1s: but SWAP_WORDS is allready defined (ARCHOS uses this)
13:28:54n1si guess it's fine then :)
13:29:46wodzn1s: about the macros - honestly I was thinking about this but I don't know how to introduce line breaks in macro definitions
13:29:47n1sS_a_i_n_t: linear scale with spectrogram isn't working here
13:30:24S_a_i_n_twell, least it's not just me ;)
13:30:33JdGordonWTF? I have the progressbar code being called, but its not drawing?
13:31:09S_a_i_n_tJdGordon: are you drawing 1 or 2 bars?
13:31:24S_a_i_n_tI've been failing to draw more than 1 bar lately
13:31:28 Join slck [0] (Venci@Slackware.SlackPix.Com)
13:31:33n1swodz: ther should be examples omewhere, i'l try to di one up
13:31:39n1sdig, even
13:32:32 Quit TheSeven (Ping timeout: 245 seconds)
13:32:43S_a_i_n_tyeah, like %?pv<mute|bar1|line level|bar2> doesn't work for me anymore...
13:32:56S_a_i_n_tremember I got you to look at it in the SIM?
13:33:03S_a_i_n_ta few days back
13:33:15 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
13:33:53JdGordonit partially worked...
13:34:09JdGordonI might have misunderstood the issue
13:34:47S_a_i_n_tyeah, the first bar drew, bot not to the full extent of its defined viewport...and the second bar just drew in full past 0dB instead of drawing like a progressbar
13:35:40S_a_i_n_tI think it failed after you fixed up the problem pixelma was having with it.
13:35:50S_a_i_n_tI never did any testing with it after that.
13:35:58S_a_i_n_tuntil the other day that is...
13:36:02JdGordonif you can hunt down the commit I can fix it quicker :)
13:36:30n1swodz: apps/plugins/mpegplayer/mption_comp_coldfire_s.S has a couple of macros that should work as examples
13:36:44S_a_i_n_tit was some problem pixelma was having with the seperators? "|"?
13:36:58S_a_i_n_tI can't remember exactly what the problem was though.
13:37:03JdGordonGRR... I tihnk I need to add a tag to say "DONT clear the text area" so text can be drawn over images
13:37:58JdGordon%pv|not|the|params| thought %pv was the long type
13:38:12wodzn1s: aaa gas macros - I was thinking about cpp macros :-)
13:39:45JdGordonamiconn: is there an LCD draw mode to do transparent but use the current lcd buffer instead of the backdrop image for the bottom layer?
13:39:47 Join adnyxo [0] (
13:40:13S_a_i_n_tJdGordon: Everything looks OK with FS #11133 installed.
13:40:18S_a_i_n_tAny specific tests?
13:40:32JdGordonall the different image tags :/
13:40:36*amiconn doesn't understand
13:40:41amiconn@ JdGordon
13:40:51JdGordonI want to draw text over an image
13:40:56JdGordonthe image not being the backdrop
13:41:27JdGordonso I'd draw the image first, the draw with a transparent backdrop and not have the backdrop show up under the text
13:42:37amiconnDRMODE_FG - that exists for ages
13:42:48S_a_i_n_tJdGordon: Your volume/progressbar isn't showing? Mine are working fine (CabbieV2 "vanilla" style, and My theme's "progressbar" style)
13:42:55*amiconn was confused by the mention of 'transparent'
13:43:53JdGordonah, thanks
13:44:11JdGordonS_a_i_n_t: yeah, its using the old code
13:44:36JdGordonI now have text drawn over the progressbar... but all other text looks shit
13:45:03*S_a_i_n_t managed text over the progressbar ages ago.
13:45:10S_a_i_n_tbut it blinked a little sometimes.
13:45:51S_a_i_n_tI did it in a pretty ugly way though
13:46:20 Nick fxb__ is now known as fxb (
13:46:50 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
13:49:56amiconnJdGordon: Btw, DRMODE_FG doesn't work for scrolling text.
13:50:17JdGordonit would look very crap if it did :)
13:50:27JdGordonI saw the effect with numbers counting
13:51:22pixelmaspeaking of drawmodes... how complicated would it be to add another parameter to viewports for "inverted" so you could have white on black on monchrome and also exchange fore- and background colours on colour displays etc.?
13:51:36pixelma@ JdGordon
13:51:46***Saving seen data "./dancer.seen"
13:52:21JdGordontechnically not at all... but doing it without breaking skins is a bit more difficult
13:52:50JdGordonOH hmm... we could use magic numbers for that
13:53:19pixelmahow? And maybe a chance for the parameters patch
13:53:35JdGordonyay! I now have text over images working nicely, but you need to add a %aZ to the line for it to work
13:54:31pixelma(the "how" was referring to the magic numbers)
13:54:36JdGordonpixelma: i dunno, so magic constants
13:55:01pixelmasorry, can't imagine what
13:55:02JdGordonthis would only be if both are set to - right?
13:55:03 Join fdinel [0] (
13:55:33pixelmawould make sense
13:57:04pixelmaof course on monochrome the colours part doesn't exist
13:57:42JdGordonyeah, ok magic values wont work so well
13:57:51JdGordonwithout breaking would be simple enough to add a new tag
13:58:09JdGordonwith breaking its add a new param
13:58:24CIA-5New commit by nls (r26100): fft plugin: move const data into a c file and use no wider types than necessary, makes the plugin about 63kB smaller
13:58:32JdGordonI wonder if it owuldnt be better to set the viewport options like font and colours in seperate tags anyway
13:59:25S_a_i_n_tthat idea seems weird to me.
13:59:27JdGordon%V|x|y|width|height then (maybe on the same line) %Vf|0| <- viewport font, %Vb|colour| <- backgroud
13:59:37linuxstbJdGordon: Why? They're attributes of the viewport.
14:00:03JdGordonyes, but it means we can add things without breaking later, and I find most of mine use defaults for font and colours anyway
14:00:13 Quit GeekShadow (Ping timeout: 252 seconds)
14:02:02JdGordonit will be more verbose, but only if you actually use those options
14:02:07JdGordonotherwise the defaults will be used
14:02:55 Quit mt (Ping timeout: 258 seconds)
14:03:15*JdGordon needs a better suggestion than %aZ for this tag which says "dont clear the text area for this line before drawing the text"
14:03:28 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
14:03:40linuxstbJdGordon: That just seems illogical to me - they're attributes of the viewport, so should be part of the viewport definition. Didn't you already add the concept of default values?
14:04:02 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier.
14:04:12JdGordonyes, but adding more things becomes difficult
14:04:20 Join teru [0] (
14:04:46JdGordon%V|0|0|-|8|-|-|-| could just as nicely be %V|0|0|-|8|
14:05:20JdGordonI'm not talking optional parameters here... actual seperate tags for the colours
14:05:56 Join aaron_ [0] (
14:06:08 Quit adnyxo (Ping timeout: 276 seconds)
14:07:33JdGordonanyway... FS #11133 would be good to get testers... if its broken images will not be drawn correctly
14:07:44JdGordonand anything "imagy"
14:07:48linuxstbHow many other things are there to add?
14:08:37JdGordon3 line selection colours (which wont probably be needed), plus any wierd options and ideas people come up with... like the above
14:08:38pixelmaI just suggested one - "inverted" but can't think of anything else
14:08:50S_a_i_n_tJdGordon: I've run all my current themes+works in progress on the FS #111133 build, and all seems well.
14:09:05S_a_i_n_tThere is quite a diverse range of tags between all my themes.
14:09:05pixelmait's not weird - at least not for monochrome
14:09:08JdGordondo the new patch please?
14:09:36S_a_i_n_tthere's another one?
14:09:46JdGordon2min ago
14:09:47n1shmm, my patch still causes crashing on the beast, seems like a nullptr deref somewhere :/
14:09:54S_a_i_n_taha...sweet, will do.
14:12:13n1sthere is an easier, more hacky way to do this of course...
14:16:11wodzn1s: FS #11283 - now better?
14:16:19 Join mt [0] (~mtee@rockbox/developer/mt)
14:18:09n1swodz: the firmware/SOURCES change looks wrong in the new patch
14:18:44n1smaybe you need to resync your repo
14:20:37wodzyes You are right
14:21:20n1sother than that i think it's ok
14:22:56wodzok so I will do final test & commit
14:23:27n1ssounds good to me
14:25:15 Join Curtman [0] (
14:29:23 Quit bieber (Ping timeout: 276 seconds)
14:29:38 Join bieber [0] (
14:30:06 Nick dys`` is now known as dys (
14:32:12 Join Schmogel [0] (
14:34:07CIA-5New commit by jethead71 (r26101): Simplify mpegplayer a bit and use array-based lists rather than linked lists for stream management. Move a couple useful functions to handle pointer ...
14:34:52 Quit kugel (Ping timeout: 252 seconds)
14:35:00CIA-5New commit by wodz (r26102): HD200 minor lcd asm routines adjustments
14:35:57pixelmaabout Rockboy adaptation to different LCDs - there's this patch but can't comment on code quality and if it would still apply
14:36:35CIA-5New commit by wodz (r26103): HD200 - decrease wait states for lcd when boosted
14:38:59 Join max242 [0] (
14:43:31 Join M3DLG [0] (~M3DLG@
14:49:53CIA-5New commit by teru (r26104): rockboy: fix format strings in snprintf. use int for slot_id and use %d tag for it in format strings.
14:50:13 Join dfkt_ [0] (~dfkt@unaffiliated/dfkt)
14:52:50 Quit dfkt (Disconnected by services)
14:52:54 Nick dfkt_ is now known as dfkt (~dfkt@unaffiliated/dfkt)
14:55:41S_a_i_n_tI don;t remember if I've asked this before or not, but is "seek" (|<< >>|) *supposed* to seek into and beyond the next track? I assumed it should reach the end of the currently playing track and go no further, but I notice on my Nanos I can hold seek down for a minute or so and end up seeking through 3~4 whole tracks.
14:55:59 Quit M3DLG (Ping timeout: 240 seconds)
14:56:03S_a_i_n_tIt would be fine if the screen updated and let you know you were on a new track, but it doesn't
14:56:15S_a_i_n_tIt only updates when you release the seek button.
15:00:39 Join vaguerant [0] (
15:00:39 Quit vaguerant (Changing host)
15:00:39 Join vaguerant [0] (~vague@wikipedia/vague-rant)
15:02:00 Part vaguerant
15:02:50 Join vaguerant0 [0] (
15:03:09 Nick vaguerant0 is now known as vaguerant (
15:03:12 Quit vaguerant (Changing host)
15:03:12 Join vaguerant [0] (~vaguerant@wikipedia/vague-rant)
15:03:25 Quit antil33t (Read error: Connection reset by peer)
15:03:31 Join antil33t [0] (
15:04:44vaguerantJust wanted to quickly ask any Clip or other AMS Sansa devs if a default keyset has been set up for the CHIP8 plugin. I've just started messing with it on my Clip+ and it seems like a few keys are "missing"; in BLINKY I can move left/right but not up/down; and yes I'm using an appropriate custom keyset.
15:07:58CIA-5New commit by wodz (r26105): Integrate byte swap of ata word into ata-as-coldfire.S
15:08:52CIA-5New commit by wodz (r26106): HD200 - use optimised read&write routines from ata-as-coldfire.S and drop hd200 specyfic ata-as-hd200.S
15:12:44vaguerantDitto MINES, the Minesweeper clone (not the plugin, the CHIP8 game); can't move up or down, only left and right. It looks like some of the buttons are just missing mappings entirely, but that's just a lay interpretation.
15:13:19n1svaguerant: you can configure your own keys in the chip8 plugin so they might just be undefined
15:13:29n1sthe manual holds the details
15:13:48n1sproblem is that the machine has more keys than any of our targets
15:14:16teruCould anyone please review or comment the patch on FS #10559 (lrcplayer: a plugin to view .lrc file) especially manual part? I want to commit it if there isn't large issue.
15:15:16Tornevaguerant: chip8 on clip+ has volup/left/select/right/voldown bound to 2/4/5/6/8
15:15:38vaguerantAhhhh, volume buttons.
15:15:45wodzwhat is the origin of random build errors as seen on
15:15:46vaguerantOf course, the one thing I stupidly didn't try.
15:16:13vaguerantYep, works perfectly.
15:16:20Tornethis is probably dumb, though..
15:17:23n1swodz: usually some issue with the server, like out of diskspace or something
15:17:33n1sgevaerts: seen the error on your server?
15:17:36vaguerantIs there a reason up/down weren't mapped to 2/8 instead of the volume buttons?
15:17:50Tornethings with c200/e200/clip pads have tht button mapping and it seems likely to be useless :)
15:18:32Tornevaguerant: not relly, no
15:18:38Torneat least, not that i know of
15:18:43pamauryyeah, I managed to enable the usb phy&core in the clip+ ! 0.01% of the work is done :)
15:19:02Tornesomeone probably copied it from another player which has volup/voldown as the up/down on the 4-way pad
15:19:25vaguerantSounds plausible.
15:19:53Tornelemme have a quick check which ones should be changed and I'll fix it
15:25:15JdGordonS_a_i_n_t: cooeey
15:25:25CIA-5New commit by torne (r26107): Change default chip8 keymap on c200/m200/clip families to use up/down on the 4-way pad instead of volume up and volume down, since most chip8 games ...
15:25:36Tornevaguerant: there you go
15:25:45 Join komputes [0] (~komputes@ubuntu/member/komputes)
15:26:33vaguerantOh great, looks awesome, will update right away.
15:28:18Tornethe pictures of the players in question show it makes sense so.. yeah :)
15:31:12 Quit max242 (Quit: CGI:IRC)
15:31:45 Join Blue_Dude [0] (~chatzilla@rockbox/developer/Blue-Dude)
15:32:18vaguerantFeels perfect, thanks Torne.
15:32:30 Quit bongo (Quit: CGI:IRC)
15:34:42 Quit wodz (Quit: Leaving)
15:36:41S_a_i_n_tWould it be possible to get the Virtual Keyboard to have a definable font?
15:37:07S_a_i_n_tswitching fonts *really* messes up the layout of a custum .kbd file.
15:37:11Torneyes, if you used the multifont support some more
15:38:15S_a_i_n_t.kbd supports multifont?!? (or is that a "yes, it's possible" thing?)
15:38:57JdGordonkbd does not support userfonts
15:39:05Tornei mean, there is a way to load multiple fonts
15:39:12Tornetherefore you could hack up the kbd code to do so
15:39:18Tornenot that this was already supported
15:39:56S_a_i_n_tJdGordon: Did you mean to say that?
15:40:03S_a_i_n_t.kbd uses the userfont
15:40:04 Join CGL [0] (~CGL@
15:40:16*JdGordon takes that back
15:40:20Torneit uses whatever font is left over
15:40:25Tornewhich is kinda broken ;)
15:40:50S_a_i_n_tIn my experience, if a .kbd file is loaded, it uses userfont.
15:40:55S_a_i_n_tif not, sysfont
15:41:12JdGordonwho do I bug about the Makefiles?
15:41:24TorneS_a_i_n_t: no, i've had it use both fonts without ever having had a .kbd
15:41:44JdGordonis there any reason we bother with appsextra in configure to find the recorder and gui folders for -I when that is static anyway?
15:41:48S_a_i_n_twow...that's messed up. I've never seen that.
15:41:49JdGordonwhy isnt it just in apps.make?
15:42:14TorneJdGordon: i think it's to not include them on Player :)
15:42:33 Join panni_ [0] (
15:42:48JdGordonbut SOURCES controls which files get compiled anyway, and #ifdefs control the .h's
15:44:23S_a_i_n_tWell, if .kbd ever gets a working over, a way to set a default font (like viewer has already) would be *great*. It'd be great to be able to change fonts and not have it mess up your custom keyboard layout.
15:45:31terucalculation of y coordinate of splash seems wrong to me. e.g. if vp.y > LCD_HEIHGT/2, splash is not shown within the ui viewport.
15:51:50***Saving seen data "./dancer.seen"
15:54:38gevaertsn1s: yes, now
15:54:59 Quit Schmogel (Read error: Connection reset by peer)
15:55:00gevaertsn1s: I claim that that's a makefile error
15:55:13n1sah, ok
15:55:34gevaertsIt looks similar to the pdbox error I fixed a few months ago
15:56:00gevaertshm, actually, I think I've seen this one before as well
15:59:26 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:59:50kugelJdGordon: maybe post pone the drawmode thing to after the skin break work?
16:00:16kugelWhy not?
16:00:20kugelis it urgent?
16:00:20JdGordonteru: very probably... feel free to fix it :)
16:00:30JdGordonit is totally unrelated
16:00:37kugelor do we desperately need *more* tags which can be done with existing ones?
16:00:59S_a_i_n_tthis won't add any tags IIUC
16:01:32 Quit mitk (Quit: Leaving)
16:01:38JdGordon1) the added tag is irrelevant to the patch (although makes it much more useful, 2) how can you draw text over images with existing tags?
16:02:44kugel%aZ is new isn't it?
16:03:10kugelI think the drawmode is a better solution and it should be part of the %V tag
16:05:12JdGordonno it isnt
16:05:23JdGordonunless you want a viewporet for every single text block to use this
16:05:35JdGordonwhich wont even work if it is conditional
16:06:28 Join mikroflops_ [0] (
16:06:29kugelthis way you need %aZ for every single *line* even
16:06:40kugeland why should it not work with conditionals?
16:07:01JdGordonconditional viewports
16:07:14JdGordonnot every line.. only the ones you want to overlap images
16:07:33kugelstill more than once per viewport?
16:07:45JdGordondepends on the skin
16:07:59kugeldon't use it as argument then
16:09:37CIA-5New commit by teru (r26108): splash.c: correct calculation of y coordinate of splash.
16:09:39JdGordondoes anyone know who is responsible for the presets code in radio.c?
16:09:45JdGordon(for the (c) notice)
16:10:59 Quit mikroflops (Ping timeout: 268 seconds)
16:11:11 Join MethoS- [0] (~clemens@
16:12:18kugelgevaerts: yes I am in schedule with the gsoc paperwork
16:15:39gevaertsgood :)
16:21:30kugelbut I think I'll write another mail asking for confirmation
16:21:39 Quit mikroflops_ (Ping timeout: 265 seconds)
16:21:56*kugel doesn't like the method of "if I don't answer it should be fine"
16:22:02 Quit JohannesSM64 (Ping timeout: 240 seconds)
16:23:07kugelgevaerts: I don't know if you know it but google messed up the SFC so they extended the deadline to friday
16:23:43gevaertskugel: I don't know details. I just know there is paperwork :)
16:24:09kugelI thought you asked because today was the original deadline :)
16:29:08 Quit CGL (Remote host closed the connection)
16:30:48 Part LinusN
16:33:00 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
16:33:38JdGordonwhich .h is IF_CNFN_NUM_() in?
16:35:54 Join JohannesSM64 [0] (
16:38:56 Quit mt (Ping timeout: 258 seconds)
16:39:21 Join mt [0] (~mtee@rockbox/developer/mt)
16:39:44 Join Xerion [0] (
16:40:24 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
16:42:51JdGordonhmm... at what point does splitting up a messy file not make sense anymore? I'm trying to pull the preset code out of radio.c but there is lots of static functions which need to be unstatic-ed, and shared globals...
16:44:43 Quit bluebrother (Disconnected by services)
16:44:46 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
16:45:43 Quit aaron_ (Remote host closed the connection)
16:45:48 Join adnyxo [0] (
16:46:38JdGordonI guess splitting means it would eventally be easier to actually fix the mess up right?
16:46:46JdGordonhow bad is un-static-ing functions?
16:46:55JdGordonand ints
16:47:19Torneif they've got sensible names that won't clash, it's fine..
16:50:27 Quit adnyxo (Ping timeout: 245 seconds)
16:52:23 Join EK [0] (
16:55:35JdGordonradio.c is almost managable now! only 1000 lines
16:57:18 Quit bieber (Ping timeout: 260 seconds)
16:57:31 Join bieber [0] (
16:57:32JdGordonB4gder: Zagor: gevaerts: you guys play with the makefiles mostly right? appextras from configure is the same for just about all targets and the differences should be handled with #ifndefs on the .h files... can we just change it to a static list in apps.make for the -I line?
16:58:31JdGordonunless there really is a issue with non radio targets having apps/radio/ in the includes search path?
17:00:07 Join M3DLG [0] (~M3DLG@
17:00:29JdGordonthe sansa view and archos player are the ONLY ones with different appextras
17:02:16JdGordonmeh :/
17:02:29Zagorno objection from me
17:02:41Zagorit's a complexity that doesn't appear necessary
17:02:51JdGordonfind+replace works fine :p next time
17:04:01CIA-5New commit by jdgordon (r26109): split the radio code into multiple files in the hope that makes it more likely for someone to want to work on! :D
17:06:47 Quit EK (Quit: CGI:IRC)
17:10:04 Quit n1s (Ping timeout: 276 seconds)
17:10:06 Quit GeekShadow (Ping timeout: 246 seconds)
17:11:11CIA-5New commit by jdgordon (r26110): fix yellow/red
17:12:01 Quit kugel (Ping timeout: 276 seconds)
17:15:43CIA-5New commit by mcuelenaere (r26111): Do s/SWAP_WORDS/ATA_SWAP_WORDS/ to prevent namespace clashes.
17:15:50JdGordon:( apparenlty I broke checkwps.make?
17:15:58 Part Zagor
17:16:32 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
17:16:45CIA-5New commit by jdgordon (r26112): global tab->spaces is not clever with Makefiles!
17:18:30 Quit Llorean (Ping timeout: 246 seconds)
17:19:14 Quit TheSeven (Ping timeout: 260 seconds)
17:20:38 Quit M3DLG (Ping timeout: 240 seconds)
17:22:35 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
17:22:53 Join M3DLG [0] (~M3DLG@
17:30:24 Join mt_ [0] (~mtee@
17:31:03 Join kugel [0] (~kugel@rockbox/developer/kugel)
17:31:41 Quit mt (Ping timeout: 240 seconds)
17:31:50 Quit M3DLG (Ping timeout: 248 seconds)
17:32:25 Join mt [0] (~mtee@rockbox/developer/mt)
17:34:28 Join funman [0] (~fun@rockbox/developer/funman)
17:35:39 Quit mt_ (Ping timeout: 258 seconds)
17:38:46funmanjust got the as3543 datasheet from AMS. asked for it yesterday through the web, got an answer this morning asking me to sign a NDA, i said "no because i'm open source"
17:39:11funmansimple and fast ..
17:39:29 Join mt_ [0] (~mtee@
17:39:58Torneso you said no and then they sent it to you anyway?
17:40:45 Quit mt (Ping timeout: 252 seconds)
17:40:55 Nick mt_ is now known as mt (~mtee@
17:41:42funmanit was a 10 lines 'no', but that's it
17:41:56funmanare you interested by the (french) mail exchange?
17:42:05Tornenot particularly, i have to say
17:42:35funmani suppose for such a fast answer they looked up my email and saw that i had previously asked other documents
17:45:17pamauryvive funman \o/
17:45:44 Quit mt (Ping timeout: 265 seconds)
17:46:33funmanit is very similar to as3517 (can't see any difference with a quick look)
17:47:20pamauryI hate the clip+ of refreshing database screen, it's sooooooooo slow
17:47:29 Quit lpereira (Quit: Leaving.)
17:49:04*S_a_i_n_t wonders where he can check out the latest revision of the "funman source" ;)
17:49:17S_a_i_n_t[03:39] <funman> ... i said "no because i'm open source"
17:49:20funmanjhMikeS: did you get the end of charge on e200v1 ?
17:49:38vaguerantpamaury: Hold power to shut down.
17:49:47pamaurydoes it works ?
17:49:51vaguerantTransfer your stuff, then, before unplugging USB, hold power.
17:50:06pamaurythat will speed up dev
17:50:07funmanbe sure to unmount before doing that (to be sure)
17:50:13vaguerantHold for like 10s; it's still like 1/10th the amount of time you have to wait for the damn database.
17:50:13pamauryyes of course
17:50:17funmanpamaury: you can also use µSD + rolo
17:50:38funmanor leave only a few files on the storage
17:51:26pamauryeven with only .rockbox it's slow. If I put it on SD, the bootloader will load the sd one first ?
17:51:54***Saving seen data "./dancer.seen"
17:51:58funmanno but you can leave the clip+ running and 'open' the rockbox.sansa with rolo
17:52:34pamauryyeah true, that's a shame I have no microsd capable reader and that I forgot (or lost) my microsd->sd converter
17:52:51funmankugel: i think we can dim the fuzev2 backlight with the hardware
17:53:41pamauryI did one of those brilliant "I put it here so I'll never lose this tiny piece of hardware" and finally forgot where was the magical place :)
17:53:50S_a_i_n_tmicroSD to USB convertors (tiny, tiny wee things...) are *very* cheap also...I got one for like ~$0.99
17:54:02 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:54:08pamauryoh, where do you buy that ? I never saw one
17:54:17vaguerantDealExtreme is cheap for those.
17:54:23vaguerantFree shipping, takes a couple weeks though.
17:54:36vaguerantIf you're somewhere western, anyway, they're in China.
17:54:37 Join phanboy4 [0] (
17:54:41 Quit teru (Quit: Quit)
17:54:46 Quit dys (Ping timeout: 276 seconds)
17:54:50S_a_i_n_tMost times I've seen them free with the microSD card.
17:54:59S_a_i_n_tThat's how I got my first one.
17:55:10S_a_i_n_teasy to lose, but easy to replace ;)
17:56:38pamauryfunman: I managed to enable the usb phy&core it seems mimicing the of code (linux now report the usb plug) but it seems the usb registers are not accessible, at least the "synopsis id" one reports 0...
17:56:51pamauryperhaps the of waits are necessary after all
17:56:53vaguerantThere's a couple <$1 on DX, e.g.:
17:57:32funmanpamaury: ids can be different
17:57:45pamauryyes but not 0
17:57:58funmanwhy not?
17:57:58vaguerantHmm, from the comments that one's not SDHC compatible, so just be careful if you pick something like that up.
17:58:07 Quit B4gder (Quit: It is time to say moo)
17:58:36pamauryfunman: sounds strange, the linux code says it must be 0xmagicxxx if it's a synopsis one
17:58:47pamaurybut I didn't check other registers however
17:59:08funmanthere's a fixed ID for AFE UID registers in linux, but in sansas they vary with each model (perhaps to be used with DRM)
18:01:14 Quit petur (Quit: Leaving)
18:01:57 Join DataGhost [0] (
18:01:57 Quit DataGhost (Changing host)
18:01:58 Join DataGhost [0] (~dataghost@unaffiliated/dataghost)
18:03:12pamauryfunman: the waits are necessary, putting waits everywhere I read0x4F54260A
18:03:35pamaury(the same as linux code)
18:03:59pamauryhowever the of waits are really long, I'm not sure there are all necessary
18:04:17funmantry reducing them / removing some
18:04:57pamauryI'm trying currently
18:07:09funmandfkt: i think the volume should be the same on all clips, i'm not sure why it's lower on v2/+ :/
18:07:12pamaurylooping 0x300 time seems a lot but removing of them doesn't work, I'll have to check them one by one
18:15:12pamauryfunman: is audio isr enabled on clip+ ? It is commented out it seems but present in the vectored interrupt list
18:16:15pamauryBut how can it pass the linking pass ?
18:16:17funmanyou have to enable it in VIC_INT_ENABLE register, and if INT_AUDIO() doesn't exist, it will be aliased to UIRQ() which just panics
18:16:35funman/weak in system-as3525.c
18:24:32 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
18:25:19 Join Schmogel [0] (
18:25:57 Join domonoky1 [0] (
18:28:34 Quit domonoky (Ping timeout: 276 seconds)
18:30:53 Quit pamaury (Ping timeout: 264 seconds)
18:31:15funmanwhen fuzev1 battery is full, adc reads as 4.129V with charger unplugged, and 4.189V with charger plugged
18:31:24funmanfull voltage is defined as 4.16V
18:32:35funmanso when end of charge is detected, the charger is disabled, battery reads now belowfull voltage, and charger is immediately re-enabled
18:35:21 Quit guysoft22 (Read error: Connection reset by peer)
18:35:38 Join guysoft22 [0] (
18:37:54 Join CGL [0] (~CGL@
18:41:34 Quit antil33t (Read error: Connection reset by peer)
18:41:41 Join antil33t [0] (
18:46:27ranmaWhy disable the charger on end-of-charge? You could just leave it enabled until it is unplugged AFAICS
18:47:17 Join M3DLG [0] (~M3DLG@
18:47:28funmanoh really?
18:47:52ranmaSure, it's in constant voltage mode then, so you can't overcharge the battery.
18:47:53funmanbattery will stop sucking current automagically ?
18:48:23 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
18:48:24ranmaSure it will continue to suck a little current, but it shouldn't be much...
18:48:52ranmaIt will supply just enough current to keep the battery at full charge.
18:48:56funmani'm sure i mentioned it already, but i'm not really friend with electricity theory
18:50:48funmanif we can skip disabling the charger and just detect plug/unplug it would make things much simpler indeed
18:51:04funmanbut jhMikeS seemed to say leaving the charger enabled would be harmful to the battery
18:52:19 Nick CGL is now known as [CGL] (~CGL@
18:52:43funmanas3514 (and newer) datasheets say:
18:52:45funman1: charging is complete, turn charger off
18:53:03funmannow it's not entirely clear if the charger is turned off, or if the datasheet orders us to turn it off
18:53:58ranmafunman: I we care for battery lifetime it would be better to not charge to 4.2V but only 4.1V or so.
18:54:20ranmaThat of course reduces the runtime of the player, but should be better for Li-Ion life
18:54:42funmanranma: then please do it, i think powermgmt-target.h is the way to go
18:57:28funmanin linux i can't find anything disabling the charger on EOC condition
18:58:07 Join GeekShado_ [0] (
18:58:39ranmajhMikeS: Why is keeping the charger running harmful to the battery? As far as I'm aware that should be safe, since it's in constant voltage mode and will just keep it at full charge, no harm done.
18:58:43funmanperhaps I could ask AMS directly
18:59:27funmanThey are "at my disposal for any further information."
18:59:35ranmaNeat :)
18:59:59funmanit doesn't mean they'd answer ;)
19:00:58 Quit GeekShadow (Ping timeout: 260 seconds)
19:01:03ranmaHmm, ok, also says 'turn off', so I guess he's right :)
19:01:51 Join stripwax [0] (
19:02:04funmanranma: the datasheet seems to imply that the charger is turned off automatically on EOC
19:02:19 Quit stripwax (Client Quit)
19:02:31ranma"No trickle charge is applied because lithium-ion is unable to absorb overcharge. A continuous trickle charge above 4.05V/cell would causes plating of metallic lithium that could lead to instabilities and compromise safety."
19:03:02 Quit M3DLG (Ping timeout: 248 seconds)
19:03:06ranmaAlso says "Charging only to 4.10V reduced the capacity by 10% but provides a longer service life."
19:04:23ranmaOn my Thinkpad I can also configure the percantage to which the battery has to fall before it gets topped up.
19:05:29ranmaSupposedly using a lower percentage and thus reducing the number of top-up charges also increases lifetime.
19:08:14kugelyea, you a li-ion should operate within (approx) 10% and 90% of its capacity for maximum life time
19:12:53 Join ssorgatem_ [0] (
19:16:36 Quit ssorgatem (Ping timeout: 265 seconds)
19:16:56funman <- new build condition for rockboy
19:17:53 Join jgarvey [0] (
19:18:46vaguerantI wonder how many GB games are actually playable in that res.
19:19:08vaguerantIs Rockboy's scaling optional, can you choose to just pan or anything?
19:19:32saratogais it really worth including rockboy on devices that don't have the screen resolution? seems like its just going to disappoint
19:19:36CIA-5New commit by kugel (r26113): Fix FS #11280 ...
19:22:36funman#ifndef IRIVER_IFP7XX_SERIES /* Temporarily disable plugins for iFP7xx */
19:22:40funmancan i remove this ?
19:24:19kugeli think so, that can be done via configure anyway
19:24:19n1sfunman: think so, it should be done in configure anyway
19:24:31funmanthanks, hivemind !
19:24:34*kugel wins \o/
19:24:47*n1s stands in the corner
19:26:21 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
19:27:58 Join Luca_S [0] (
19:29:07 Quit GeekShado_ (Ping timeout: 258 seconds)
19:29:17jhMikeSranma: read up on lithium ion batteries. it's not nicd.
19:29:54jhMikeSthey should shut off after the current drops. really, there no reason the charger shouldn't
19:30:17ranmajhMikeS: I did, yeah it isn't
19:30:47 Part watto
19:32:19funmanhm why splitedit has samsung yh and iriverh100 keymaps if it's only built on hwcodec ??
19:32:41jhMikeSthe problem is you have a read in a true ISR, then the read of the status in the charging code. I'm sure the ISR would steal the flag before the charging code sees it.
19:32:52pixelman1s, kugel, funman: some plugins are even prepared for the ifp though (resolution is the same as Clip etc. so could be reused but I even saw some keymaps and once build a sim to try out my graphics changes for Rockblox... but the port seems to have been abandoned :\
19:32:54jhMikeSam I correct?
19:33:22funmanjhMikeS: well it works fine on as3525 with isr
19:33:38 Join Jaykay [0] (
19:33:57vaguerantThere's an example of how Game Boy games might very occasionally be slightly less than completely horrible.
19:34:07BeFaloufor what you say, I shouldn't be charging my fuze within rockbox, should I?
19:34:20vaguerant(Framed is a 128*64 viewport.)
19:34:30funmanjhMikeS: did my diff work ?
19:34:55funmanBeFalou: if you use a current build you shouldn't be charging any sansa
19:35:20jhMikeSfunman: 1) can you elaborate 2) e200v1 charge ended at 4.212V after about 50min after levelling
19:36:05funmanjhMikeS: 2) after levelling -> this means you got EOC signalled 50 minutes after reaching max voltage ?
19:36:22jhMikeSfunman: I'm sure the ascodec ISR works fine. it's what it might be doing to the non-ISR readouts of status?
19:36:33funmanwell my goal is to remove them
19:37:04funmanhence replacing all the reads by functions which will use the bits read frm the isr on as3525
19:37:07ranmafunman: Maybe 'end of charge' is a 'level' type irq
19:37:18jhMikeSfunman: it's not a brick wall and it continues to rise a little, so after "level off" means, "after hitting voltage regulation"
19:37:21funmandatasheet says edge
19:37:33 Join watto [0] (~watto@
19:38:02jhMikeSit's edge and that's why it gets missed that way (didn't even need the DS to guess that) :)
19:38:44funmanjhMikeS: if the battery level is below 4.16V when the charger is disabled then it will be re-enabled
19:39:04jhMikeSfunman: yes, that was fine too (replug)
19:39:24funmanjhMikeS: what's "hitting voltage regulation" ?
19:39:35ranmaIs there a reason to have different voltages for first plugin and top-up?
19:39:55jhMikeSbut pp sansas have no status readout conflict anywhere. I wouldn't change it for them unless the as3514 int gets enabled properly. I had an inkling in the port pins of which it is.
19:40:02ranmas@voltages@voltage thresholds@
19:40:42funmanjhMikeS: if charging got disabled even after unplug replug, that means the patch is ok. what do you think of committing it ?
19:40:52jhMikeSranma: to avoid excessive recharge after settling a bit. could possibly raise it a bit.
19:41:20 Quit swilde (Remote host closed the connection)
19:41:24funmanthis just show we can enable interrupt monitoring once for all, and avoid writing the interrupt register multiple times
19:41:52jhMikeSfunman: for a v1, it seems fine. I unplugged and replugged while it was still low enough. it should restart again if it's not topped.
19:42:26funmani'll do that later
19:42:46funmannext step is to cache ENRD0 reading and have functions which read & acknowledge individual bits
19:42:49ranmajhMikeS: No, I didn't think of raising it, rather the other way around, reducing the 'first plugin' threshold to the same level :)
19:44:24ranmaI just don't see the point in having two thresholds there, for one I'd like to set it so that it won't start charging unless the battery level is <80% or so for me personally. :)
19:45:31jhMikeSfull down to equal vauto? then it would enable and disable the charger in quick succession. it needs the hysteresis.
19:46:31jhMikeSranma: well, do whatever on your own device :) it's not meant to work that way.
19:47:13*jhMikeS think people think these things are done just for the hell of it rather than after much research on the topic
19:49:36 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
19:50:24*ranma just wants to prolong battery life by charging to 4.1V instead of 4.2V and using a lower top-up/plugin threshold :)
19:51:46n1spixelma: yeah, it hasn't seen any changes in years but anyway, port specific defines shouldn't clutter the code like that
19:51:56***Saving seen data "./dancer.seen"
19:52:11jhMikeSactually, I don't think it's a problem. once it's activated, it charges and then waits for it to drop. if you want to increase life, I'd say to lower the regulator voltage, not the start voltage. as follows: auto <= plugin < regulator
19:52:42ranmaThat's what I said, lower the regulator voltage to 4.1V :)
19:53:13jhMikeSI thought you meant the start voltage after plugin
19:57:11 Quit GeekShadow (Read error: Connection reset by peer)
19:57:39 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
19:59:08ranmaThat would be in addition to that to reduce the number of top-up cycles. (If that has an effect on lifetime)
20:02:32ranmaFor my developer device I'm more concerned about the plug-in threshold being higher than VAUTO_RECHARGE, with the current 4.16V setting it probably enables the charger on almost every plug-in, I think I'll locally reduce both to 4.0V :)
20:03:22ranma(Given that a fully charged battery seems to be about 4.16V on the ADC here...)
20:04:40 Join GeekShado_ [0] (
20:05:03kugelgevaerts: finally got around updating the wiki page, will now write my weekly status update
20:05:04 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
20:05:34jhMikeSranma: no, it doesn't. you have to run it a bit.
20:06:21ranmaWell, it does according to my battery logs.
20:06:35 Nick fxb is now known as fxb__ (
20:06:57jhMikeSI guess it depends on what "almost every time" means. it's certainly shouldn't be so long as it finished.
20:07:04gevaertskugel: did you see Blue_Dude's questions last night?
20:07:05ranmaIt was fully charged in OF (charging stopped), then I rebooted into rockbox and immediately started the battery bench
20:07:24ranmaAnd the first logged values is about 4.15V
20:07:44kugelgevaerts: no, I'll have a look
20:07:56 Quit GeekShadow (Ping timeout: 240 seconds)
20:08:08jhMikeSranma: maybe the OF uses a lower V?
20:08:36ranmaWhich is well within the charger tolerance for 4.2V end volatage AFAICS.
20:08:52kugelpart of the problem is that a simulator build doesn't really define DEBUG (which we always thought it did)
20:09:10ranmaDatasheet says tolerance is -50mV +30mV from nominal
20:09:38ranmaSo with 4.2V nominal 4.15V is within that.
20:10:24 Nick fxb__ is now known as fxb (
20:11:18jhMikeSwhich datasheet?
20:11:55ranmaPage 153, Table 139
20:13:31jhMikeSI guess that seems right ~1%-ish tolerance
20:15:23 Quit [CGL] (Remote host closed the connection)
20:16:39jhMikeSI've never seen any go on the low side though...lift always seems to pull above the regulator setting and after some settling, around 4.17 or so
20:16:49CIA-5New commit by funman (r26114): plugins: reorder SOURCES conditions to be more readable ...
20:20:21 Quit GeekShado_ (Ping timeout: 248 seconds)
20:21:08ranmaAdd to that the 1% tolerance of the ADC and your worst case is about 4.11V measured at charge end with a 4.2V setting :)
20:21:55gevaertskugel: would it be hard to have an advanced setting in configure to switch between native libc and ours?
20:22:13kugelgevaerts: not at all
20:23:20gevaertsMaybe that would be a good way to keep people happy
20:23:20kugelthe question is just whether to only toggle the .c files or the headers too
20:23:48gevaertsI'd do both, mostly because I'm a but wary of weirdnesses otherwise
20:25:38jhMikeSranma: even if it really ever turns out that way, I don't think it should be geared toward the worst possible case since it's improbable, not typical.
20:25:39kugelI think it would be advantageous to keep the system's headers, to check whether we still follow the standard
20:25:39 Join evilnick|ipad [0] (
20:25:39kugelit shouldn't cause weirdness
20:26:31kugelalso, we don't debug headers :)
20:26:31gevaertsNo, that's hopeless :)
20:26:31jhMikeSranma: on e200v1, OF just seem to start the charger and lets it run 'till it's charged. heck I think they even set a higher V or run longer past EOCH since the battery V is a bit higher out of OF.
20:26:33 Quit Luca_S (Quit: CGI:IRC (Ping timeout))
20:27:45 Join Luca_S1 [0] (
20:27:52 Nick Luca_S1 is now known as Luca_S (
20:28:07 Quit DerPapst (Quit: Leaving.)
20:33:51 Join DerPapst [0] (
20:37:07 Quit vaguerant (Quit: Java user signed off)
20:37:57 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
20:45:01 Quit flydutch (Quit: /* empty */)
20:54:48 Join tenfoot [0] (
20:56:46tenfootfunman: are the audioset registers in the as3543 the same as those in the as3527 manual?
20:57:54tenfootthe as3527 seems quite different from what's used in the rockbox code. I'm trying to get line out working
20:58:58funmanwhich register are you looking at particularly ? i can compare the 2
20:59:14 Join pamaury [0] (~c2c7a50a@rockbox/developer/pamaury)
21:00:12tenfootmainly the audioset1 & audioset2 - according to the as3527 manual, the DAC on and SUM on are in audioset2, but the rockbox code is using audioset1
21:00:56funmanusb detection works on clip+
21:01:09 Nick YPSY is now known as Ypsy (
21:01:18tenfootI tried to program it according to the as3527 manual and all I got was silience (I was just trying to keep the headphone out working before starting on the line out)
21:03:06funmanthe audioset1 is different on all
21:03:43funmanwhat a mess :/
21:04:48funmanwe set what we think to be DAC_on (bit 5) and SUM_on (bit 6)
21:05:08funmanon as3543 DAC_on is bit 6, so it's not the same but we set it anyway
21:05:29funmanand bit 5 is DAC gainstage power (1 = on, needed for playback mixer)
21:06:13tenfootso where are the line out on bit(s)
21:06:21 Join max242 [0] (
21:07:09funmanas3514.c: 209 -> replace (0<<7) by (1<<7)
21:07:14funmanit should select between headphones & line out
21:08:28funmanLINE_OUT_* registers are not present anymore
21:08:51max242hey, i just read following post:
21:08:51 Part watto
21:09:20max242this looks very similar to what i experience with PictureFlow
21:09:37funmani had my fuzev1 crash 2 or 3 times today
21:09:38max242which i did report to some of you guys yesterday evening
21:09:43jhMikeSfunman: any more? like they got raptured, or 3543 doesn't have them? :o)
21:09:53pamaurydoes rockbox hide files beginning with a "." by default ?
21:09:55tenfootOK - I'll give that a try. Is the volume also set from the same register as the headphones or are there still separate ones for that?
21:10:37linuxstbkugel: Any idea why the sim isn't using -DDEBUG any more? It definitely used to...
21:11:44max242i don't think it is a coincidence that also PictureFlow stopped working since 25298
21:11:56funmanjhMikeS: they are marked as 'reserved'
21:11:58*jhMikeS keeps trying to find the beasts LCD panel init code and has so far failed to locate it (and is thus not all here and very irritable) :\
21:12:09max242is bor_ka also around on irc every now and then?
21:12:13funmantenfoot: yes it should be the same than headphones
21:12:23funmanafair they removed all double registers
21:12:38funmanline1 & line2 -> line; mic1 & mic2 -> mic, lineout & headphones -> headphones
21:12:54funmanand they added a bit to select between line1/line2; another for lineout/headphones ..
21:12:59jhMikeSfunman: I see that same thing with WM codecs. They have a few "classes" that are similar.
21:14:01max242and since the database/pictureflow lockup started at 25299, i think someone knowing the code could maybe pinpoint what's the cause of these lockup troubles on the fuze v1 since 25299
21:14:16funmanpamaury: do you have an usb driver that builds?
21:14:36jhMikeSfunman: there is one problem though is that I'm not sure the PP i2c even has an int, and the current i2c code is thread-only (no async for sure).
21:14:46pamauryfunman: something that build yes, call it a driver if you want
21:14:51funmanmax242: hmm thanks for pointing to this commit
21:15:10funmanpamaury: i'm going to enable usb stack for clip+ since usb detection works
21:15:17funmanand it needs CONFIG_USBOTG defined to something
21:15:40pamauryI defined it to USB_AS2525v2 and defined USB_AS... to 2535
21:15:53max242funman: well, you aksed me to pesker all you developers on irc, and so here i am
21:16:02pamauryfunman: does it really works ? Reliably I mean,
21:16:04funmanmax242: cheers :D
21:16:11funmanpamaury: yep
21:16:25pamauryfunman: there aren't thousand of audio int anymore ?
21:16:31funman100% of the one time i tried :D
21:16:38funmani'm not using interrupts
21:16:47funmani centralize reads to one place
21:16:48pamauryHow do you ?
21:16:52jhMikeShrm...damn, it's there in pp5020.h I2C_MASK
21:17:11funmanright now i'm trying to think of a better patch
21:17:15pamauryok, if you say it works I believe it :)
21:17:19 Join M3DLG [0] (~M3DLG@
21:17:31max242i really recognize what is indicated in FS #11267 also as the trouble with pictureflow
21:18:14max242although i could not see nor find a relation with cue sheets being enabled
21:18:18 Quit JohannesSM64 (Quit: WeeChat 0.3.3-dev)
21:18:55pamauryfunman: do you have a patch for usb detection ? I can try it during my experiments on usb
21:18:57kugellinuxstb: I found how to fix it at least
21:19:00tenfootyep - the line out works now. Just need to figure out which GPIO is used to detect the dock....
21:19:23 Quit DerPapst (Quit: Leaving.)
21:19:53jhMikeSfunman: the info is already gathered in ascodec-as3525.c, right? so I'm confused about all the difficulties.
21:20:23 Join petur [0] (~petur@rockbox/developer/petur)
21:20:50 Quit GeekShadow (Ping timeout: 245 seconds)
21:20:52funman <- works (except clip+ doesn't build with CONFIG_USBOTG undefined)
21:21:01 Join Peter93 [0] (
21:21:06funmansorry it's imcomplete
21:21:06 Quit Luca_S (Quit: CGI:IRC (EOF))
21:21:27linuxstbkugel: Fix what?
21:21:45kugelthe simulator DEBUG thing
21:22:13pamauryfunman: do you keep #define USB_HANDLED_BY_OF ?
21:22:14jhMikeSeverything that reincarnates is incomplete
21:22:23pamauryfunman: do you keep #define USB_HANDLED_BY_OF ?
21:22:31pamauryoops, bug :)
21:24:13 Quit komputes (Ping timeout: 260 seconds)
21:24:21Peter93Hi guys, is there any scripting support in rockbox? I want to copy or create a file automaticaly at startup...
21:24:22linuxstbkugel: What's the problem?
21:24:25funmanthis just disables the code that check if rockbox binary has been updated and ask for reboot
21:24:35n1sPeter93: not really
21:24:53funmanif you update rockbox.sansa/ipod/whatever when using rockbox usb, it asks if you want to reboot on unplug
21:24:58n1syou could write a plugin that runs on startup and does ti though
21:25:28 Quit Schmogel (Read error: Connection reset by peer)
21:25:42Peter93hm yes
21:25:43kugellinuxstb: -DDEBUG is in the wrong env variable in the Makefile, so either looking at it got lost or I don't know
21:26:00Peter93i'll try
21:27:20max242i also do notice that the PictureFlow freeze issue is not reproducable in the UIsimulator, i tested this 10 times in a row, and never ran into freeze issues
21:27:23linuxstbkugel: That was my question - why is it like that? i.e. was it just an oversight (e.g. in Zagor's makefile rewrite), or was it intentional?
21:27:30tmztfunman: how does it check that on a usb storage connection? just by monitoring the filesystem?
21:27:44kugellinuxstb: i have no idea...
21:27:46Peter93n1s: it would be a rly nice feature... what do you think about it?
21:27:48pamauryfunman: indeed, it works for me, that's great
21:28:28funmantmzt: see apps/misc.c
21:28:39n1sPeter93: it's not something i would use personally, why do you need to copy a file on startup?
21:30:09 Quit M3DLG (Ping timeout: 240 seconds)
21:30:20linuxstbkugel: Then how can you "fix" it?
21:30:35 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
21:31:41jhMikeSweird, the ascodec I2C handling seems quite like the SPI for imx31 even though they were done independently
21:32:09kugellinuxstb: by moving -DDEBUG to an env that's not ignored, or by unignoring the DEBUG env variable
21:33:52Peter93n1s: I like to autorecreate a autorun.inf file. (label & icon for Windows)
21:34:43Peter93n1s: or even desktop.ini files for folder icons...
21:34:50n1sPeter93: why do you need to do that on every startup? if you don't delete the files they should remain in place...
21:35:42Peter93n1s: if i del my mp3s or so...
21:36:06 Join Schmogel [0] (
21:36:08linuxstbkugel: You're missing my point. I'm simply asking if it's a bug that DEBUG isn't set for simulators, or if someone changed that on purpose, but left that environment variable there, unused.
21:36:21funmanpamaury: ok, please commit your work so I can just use CONFIG_USBOTG USBOTG_AS3525v2
21:36:34pamauryI can't commit, I behind a proxy :(
21:36:38funmanI just need to make a good comment
21:36:39funmanoh :/
21:37:03n1sPeter93: well, as ia said it should be fairly easy to make such a plugin, and enable the AUTOROCK feature to make it auto run
21:37:08pamauryI can send you a patch with my work+your work if you want
21:37:49pamauryOh you wait until tomorrow morning :)
21:38:03kugellinuxstb: again, I have no idea
21:38:09Peter93n1s: yeah would be rly nice...
21:38:14funmanwell i can wait, i'll just use USBOTG_AS3525 even if it's incorrect
21:38:18kugelI'm not missing the point but I already answered that
21:38:33jhMikeSthe datasheet for the that controller was obtained?
21:39:00funmanjhMikeS: USB controller of clip+/as3525v2 ?
21:39:02kugelI'm just answering your questions
21:39:49kugelI strongly suspect it's a bug since it used to define DEBUG and everybody thought it still does
21:39:55jhMikeSfunman: v2 is a different matter then than 3525v1?
21:40:02 Quit sinthetek (Read error: Operation timed out)
21:40:15pamauryjhMikeS: yes
21:40:27pamaury(for usb)
21:40:55funmanit uses as3525 base but with lot of stuffs from as353x (the CPU, the SD controller, the as3543 AFE, the USB controller)
21:40:57 Quit hebz0rl (Ping timeout: 276 seconds)
21:41:06funmanthe IRAM also i think since it's bigger
21:41:41jhMikeSso on v2 all the modules have obtainable docs?
21:41:53FlynDicetenfoot: This forum topic may be of some use to you:
21:42:02linuxstbkugel: Yes, I have the same suspicion, but I don't know if there's another reason DEBUG isn't defined, so wouldn't try to "fix" it without investigating when/why it was removed.
21:42:20Peter93n1s: I'm not so advanced in coding...^^
21:42:23kugelwhich is why I haven't fixed it yet :)
21:42:57pamauryjhMikeS: depends on what you call doc...
21:44:03jhMikeSpamaury: some writeup that tells you what the registers are, what bits are, and how to use them.
21:44:19n1sPeter93: then this is an opportunity to learn :)
21:44:26pamauryjhMikeS: then partially yes, for usb
21:44:45 Join toffe82 [0] (~chatzilla@
21:47:36max242i justed wanted to post a comment on a FlySpray ticket, and i did want to register
21:47:39 Join komputes [0] (~komputes@ubuntu/member/komputes)
21:47:48max242but apperently i do need a jabber-ID?
21:47:56funmanno it's not necessary
21:48:02max242i wanted to leave it emptye
21:48:09max242but that did not work
21:48:12funmanjust do it, jabber notifications don't work anyway afaik
21:49:10max242hmm, i'm confused, but i'll check my e-mails first
21:49:48 Quit kugel (Remote host closed the connection)
21:51:00max242nope, there's nothing in my inbox, although i get an error message the 'e-mail address or jabber-id already has been taken' when i press the 'send code button'
21:51:37max242is there maybe something wrong with registering new users at flyspray?
21:51:50linuxstbmax242: I think you'll need to wait for Bagder or Zagor to be around - they administer flyspray.
21:51:57***Saving seen data "./dancer.seen"
21:52:14max242ok, no problem, i can wait
21:52:18linuxstbAFAIK there are no known problems...
21:52:25funmanhmm clip+ can't reboot to OF :/
21:52:50 Join DerPapst [0] (
21:53:03linuxstbfunman: Do you mean rolo, or dual-booting after a reboot?
21:53:19 Quit Peter93 (Ping timeout: 264 seconds)
21:54:16 Join kugel [0] (~kugel@rockbox/developer/kugel)
21:56:16funman2nd one, we don't rolo OF
21:56:16 Quit max242 (Quit: CGI:IRC (EOF))
21:57:22pamaurywhat are the different option to sleep in rockbox ? Is there only sleep(ticks) ?
22:00:12 Join lpereira [0] (~lucien@
22:00:22 Join merbanan [0] (
22:00:23kugelget_action, queue_wait
22:00:54kugelwakeup_wait too
22:01:51 Quit kugel (Remote host closed the connection)
22:04:00 Join Dobson [0] (~Administr@
22:05:40 Join Peter93 [0] (
22:06:37funmanhm clip+ alternates between charging & discharging, checking if it's my fault
22:08:01 Join Strife89 [0] (
22:08:18 Join Blue_Dude [0] (~chatzilla@rockbox/developer/Blue-Dude)
22:08:22 Join kugel [0] (~kugel@rockbox/developer/kugel)
22:08:42 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
22:10:08funmanno it was already like that (perhaps the reason why it doesn't charge fully)
22:10:16Blue_Dudekugel: There were several changes made in "copying" the SDL code in r26065. Those changes reverted and/or added code that breaks sim function. Do you know what's going on there, and do you have a fix?
22:10:20pamaurythe clip+ seems to have 6 endpoints, and there are no equivalent, nor symmetric from IN/OUT pov, I'm checking that...
22:10:38Dobsonclip+ crashes a lot on recording
22:10:47funmanDobson: which message ?
22:10:56Dobsonfew of them
22:11:00Dobsonone is panic
22:11:10funmanplease give the full message
22:11:13kugelBlue_Dude: what sim function are you refering to?
22:11:21Dobsoni cant tell you it know
22:11:23kugelanything else besides the debugaudio?
22:11:39funmanDobson: then write it down next time it happens
22:11:40Dobsonanother crach is when player goes blank
22:11:43bertrikFS #11284 and FS #11285 have translations for Afrikaans and French done with google translate
22:12:36Dobsonwhy there is no sansa fuze v2 in rockbox utility?
22:12:55funmanbecause it was added after last release of rockbox utility
22:13:08bertrikI don't like it, but I can't really tell how good the translation is. What do other people think of these kinds of translations?
22:14:37Dobsonso what should i do if i want to install fuze v2 ( player has no screen and buttons....)
22:14:37funmanpure crap
22:14:49funmanDobson: wait for next rockbox utility release
22:14:52funmanbertrik: ^
22:15:14pamaurybertrik: completly crappy
22:15:17 Quit dfkt (Quit: -= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.)
22:15:29Dobsonif I will have time i could try to translate to lithuanian
22:15:32pamauryhehe, "la peau écran de la radio"
22:15:33Blue_Dudekugel: well, there are conditional defines in there. Maybe they are "always" defined, but in that case why have them?
22:15:54Blue_Dudekugel: And also why revert the changes I made a couple of weeks ago?
22:16:24*jhMikeS sees another extra thread creeping back into SDL :)
22:16:25Blue_DudeI guess you weren't using the cirrent version when you copied it over.
22:17:04pamauryfunman: your patch does not detect usb if it's connected at startup
22:17:22funmancharger is detected?
22:17:43pamauryno, I don't see the little icon
22:18:00bertrikfunman, pamaury Ok thanks for your opinions :)
22:18:04funmanworks for me
22:18:15Blue_DudejhMikeS: Well, yeah, I'm trying to keep my own changes in sync with the file move and some commits vanished. I'm not sure what else changed.
22:18:16 Quit toffe82 (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
22:18:35pamauryreally, I rolo with usb connected and don't have the charging icon nor get a usb connected event
22:18:41Dobsonwhere can I get pinout's of buttons of sansa fuze v2?
22:18:58funmanpamaury: what if you start with power off instead of rolo ?
22:19:35pamauryhey, but my usb code doesn't detect the controller correctly then :(
22:19:56 Quit FlynDice (Remote host closed the connection)
22:20:02kugelBlue_Dude: which changes?
22:20:06funmanperhaps because there isn't a change in usb connection
22:20:10kugelmaybe git merge failed
22:20:11*jhMikeS was speaking of r26113 (bring back the preemptive thread)
22:20:28kugeljhMikeS: is that a problem?
22:20:35pamauryhum wait, if I unplug, it doesn't detect unplug. It might be that my usb code is broken also
22:20:47kugelI told you it can survive :)
22:20:49tmztfunman: had to update and ignore conflicts, you mean check_bootfile which is called from the usb code in apps/misc.c ?
22:20:49jhMikeSkugel: no, I guess it just wants to be there
22:20:55kugelI wasn't aware it must because of windows :P
22:21:06funmantmzt: i think yes (don't know this code)
22:21:13funmani just grepped for HANDLED_BY
22:21:36jhMikeSkugel: probably has to be that way because a queue head can only be used by the thread that it was created for on Windows
22:21:43kugeljhMikeS: I think it's ok to have that, it works pretty well. and it should now work with gnu pth too which it didn't before
22:22:20kugeljhMikeS: yea, I think the SDL docs say something similar
22:22:45kugeljust strange that creating the queue head is done in SDL_INIT_VIDEO..
22:22:47 Quit Schmogel (Read error: Connection reset by peer)
22:23:01pamauryfunman: sorry, my mistake, I used the wrong firmware file :)
22:23:16jhMikeSkugel: I have noticed however, that buttons seem dodgey. Sometimes if repeating, it stops while pressed and sometimes it gets a huge flurry of buttons. It was crisp before (running debian VMWare).
22:23:21Blue_Dudekugel: The changes are here:
22:23:31Blue_DudejhMikeS: sorry.
22:23:37pamauryI'm a bit puzzled by the usb hardware config, it reports EP0 as being neither in nor out capable...
22:24:05pamauryperhaps it's special for ep0
22:24:06kugelBlue_Dude: that's my change. I was asking which of your changes you're talking about
22:24:59Blue_DudeNo, *your* changed didn't just move the file, it reverted it to an old state. Took out some things that someone else did too. jhMikeS perhaps?
22:25:18kugelBlue_Dude: I think I made this change before I went over to use the host system's C library. in the target tree there's no fprintf available
22:25:34*jhMikeS needs bringing up to speed on these changes?
22:25:40kugelthat DEBUG isn't defined for the sim is believed to be a bug
22:26:01kugelBlue_Dude: "Blue_Dude> kugel: And also why revert the changes I made a couple of weeks ago?"
22:26:09Blue_Dudekugel: You also added some #ifdef DEBUG lines that change its function. They weren't there before.
22:26:16funmanooh charger register is different on as3543
22:27:07Blue_Dudekugel: I last changed sound.c here:
22:27:39 Join Schmogel [0] (
22:27:58kugelthat was a mistake if I reverted that
22:28:25kugelI probably didn't notice this subtle changes when git reported merge failure
22:28:31Blue_Dudekugel: jhMikeS last changed sound.c here: and
22:29:05 Quit bieber (Ping timeout: 260 seconds)
22:29:22 Join bieber [0] (
22:30:12kugelBlue_Dude: jhMikeS' changes are in
22:30:14jhMikeSa few dummy functions for recording and hardware, nothing really functional
22:30:51 Quit Jaykay (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
22:33:00 Quit tenfoot (Quit: CGI:IRC (Ping timeout))
22:34:35 Quit bieber (Ping timeout: 240 seconds)
22:34:35 Join bieber [0] (
22:36:38 Quit Peter93 (Ping timeout: 246 seconds)
22:37:37jhMikeSkugel must be setting a record for moving files around
22:37:47CIA-5New commit by alle (r26115): Fix convbdf: now .c and .fnt files can be generated in one run again
22:37:52Blue_DudejhMikeS: we probably ought to either preserve files or change them. I'd rather not set the precedent of moving them *and* changing them without a good reason.
22:38:07Blue_DudeThat's why I made a point of including your commits.
22:39:10jhMikeSBlue_Dude: could be right when commits say "197 MORE files...". besides, I looked and they just got moved into target/hosted/sdl
22:39:42Blue_DudeIt makes me nervous about what else was moved and unintentionally changed.
22:41:03*pamaury just finished dumping the 4 hardware config registers of the usb clip+
22:41:18funmanpamaury: i think usb unplug isn't working fine because we rely on power thread to detect plugs/unplugs
22:41:38jhMikeSexcuse me, they're there in firmware/drivers/audio/sdl.c
22:41:40pamauryfunman: I was wrong, it seems to works, I used a old firmwae when power off/on
22:42:07funmannow charging should work alright too
22:43:37jhMikeSeh? it discharged instead?
22:43:37funmanis the power thread still running in usb mode ?
22:43:46jhMikeSfunman: it's always running
22:43:50funmanjhMikeS: i don't know what it did
22:45:06Blue_DudejhMikeS: I'm snakebit man. Every time I try to be helpful I end up making it worse.
22:45:06funmancharger register is at a different place so i have no idea why it charged anyway
22:45:06funmanbut it stopped at 67%
22:45:06pamauryfunman: it's probably waiting for disconnect :)
22:45:06funmanwell disconnecting still leaves the usb screen
22:45:06DBUGSent KICK funman to server
22:45:06funmanlet's ignroe that for now
22:45:15Kick(#rockbox funman :No flooding!) by logbot!
22:45:28 Join funman [0] (~fun@rockbox/developer/funman)
22:45:30pamauryhe, why logbot did that ?
22:45:34*pamaury loves when it works but nobody understands why
22:45:38jhMikeSBlue_Dude: understood. Usually I end up creating confusion or being confused myself. :)
22:45:59funmanperhaps it can charge slowly even with the charger disabled?
22:46:11funmani noticed the fuzev1 charged much faster than the v2
22:46:36AlexPlogbot seems to have got very trigger happy recently
22:46:40jhMikeSfunman: not that I ever noticed. it disconnects the battery FET.
22:47:12jhMikeSor perhaps, better to say, turns it "off". don't know how it would disconnect it. :)
22:47:54jhMikeSyou might get a voltage rise just from that battery having the load removed
22:48:30kisakgood afternoon
22:48:51kisaklet's see what fun we have today *reads shortlog*
22:49:15jhMikeSkisak: no fun, sorry, except for funman
22:49:32kisakand there's a problem with that?
22:49:54kugelBlue_Dude: you're changes are missing, do you want me to restore them or want to do it yourself?
22:50:17AlexPI like how the machine translations replace good translations with bad
22:50:48kugel"<jhMikeS> kugel must be setting a record for moving files around" ?
22:50:49AlexPI wouldn't dream of attempting to translate the French and even I can tell they are rubbish :)
22:51:11 Join M3DLG [0] (~M3DLG@
22:53:13jhMikeSkugel: I just haven't seen so much shuffling so far in SVN except with this hosted conversion
22:53:28CIA-5New commit by funman (r26116): as3514/as3543 fixes ...
22:54:40 Quit polobricolo (Ping timeout: 240 seconds)
22:54:48kugeljhMikeS: yea, quite a job
22:55:32funmanhm haxx have a problem ? :/
22:55:57CIA-5New commit by alle (r26117): Improve spacing in comments in convbdf; rename a local variable which can be confused with 1 (one)
22:56:57pixelmanow a cygwin sim doesn't build at all for me /home/Marianne/SVN/rockbox/firmware/target/hosted/sdl/system-sdl.c:80: error: parse error before "cond_signal"
22:58:55pixelmaand more like that
22:59:19Blue_Dudekugel: to be honest I've done quite enough damage already. If logf works in the new location, It would be nice if you could restore them. Sorry for the bad advice re: jhMikeS's changes. I wasn't aware of the file split.
23:00:37kugelpixelma: meh, can you try adding "#include <inttypes.h>" somewhere at the top near the other #includes?
23:00:41Blue_Dudekugel: If logf doesn't work, then I'll put in a local DEBUGF myself and don't worry about it. The functions being made static would be nice though.
23:00:52Blue_DudeAs long as they work with the file split.
23:01:03kugelI wonder why to logf in a sim code?
23:01:23Bagderto use the same code?
23:02:00Blue_DudeConditional "debug". One define and you turn off the debugs for everyone expect whoever needs them.
23:03:28Blue_DudeNot strictly useful if the same code will appear in RaaA, but useful otherwise.
23:04:03kugelit ought to be as useful as in a dap build in RaaA I think :)
23:04:13 Quit Schmogel (Ping timeout: 265 seconds)
23:04:35 Quit S_a_i_n_t (Ping timeout: 245 seconds)
23:04:36*pamaury thinks the usb hardware of the clip+ has too many registers, that's horrible
23:05:09funmanBagder: my svn ci was locked and one of the last build rounds had a problem
23:05:29Bagder2 k jte
23:05:56Blue_Dudekugel: it might at that.
23:06:01Bagder2 mins before your comment you committed, 1 minute after someone else committed. and I was connected to the svn master over ssh during the entire time...
23:06:02gevaertsBagder: if you typed that on the same machine, it could explain things :)
23:06:21 Join fml [0] (
23:06:30pixelmakugel: in system-sdl.c?
23:06:37kugelpixelma: yes
23:06:54fmlkugel: hello. You found statement rearranging confusing?
23:07:05funmanranma: ping
23:07:14kugelpixelma: if not, s/uint32_t/Uint32/ should do the trick
23:07:20fmlpixelma: I'll stay here for about 20 mins and we can discuss FM frequency displaying
23:07:48kugelfml: yes, because there's no clear definition of what % does to negative numbers
23:08:33kugelI wonder about the advantage of the rearrangement, I suspect the binsize reduction (which was more or less non-existent) resulted from the while -> if
23:08:38fmlkugel: I think there is. But the code works with both possible definitions.
23:09:39fmlkugel: yes, while -> if. And the code also potentially runs faster (less iterations needed). Although in this particular case this won't be seen.
23:10:35 Quit n1s (Quit: Lmnar)
23:10:48kugelI would appreciate if the % was moved back down the if () where its result is very clear
23:11:02funmanpamaury: it should detect USB correctly when rebooting
23:11:32pamauryfunman: ok, you plan to commit it anyway ? I'll wait until tomorrow so I can commit my work and update my tree
23:12:01funmanpamaury: you mean commit r26116 ?
23:12:20pixelmafml: I didn't understand your reasoning at all yesterday. The "precision" I'm able to adjust (or the tuner seems to do internally) is 0.05 MHz and there is a mismatch in what is displayed and how it seems to work from what I can see as an end user. I'd like to know which of the two 91.5 MHz autosaved presets is the wrong one.
23:12:25pamauryno your last pastebin
23:12:39funmani didn't test it
23:12:50funmani want to have a look at mkamsboot
23:12:51pamauryah :) I can check that
23:13:13funmanbut i hesitate, i'm tired and i don't want to make bricking mistakes
23:13:37funmanranma's code for c200v2 dualboot read usb status directly without setting the interrupt
23:13:54funmani guess the status is always valid but we only update it on a transition, if it boots with usb there's no transition
23:13:55pixelmafml: actually, what I didn't like about this commit is added code and complexity to suppress something I didn't mind at all seeing and which was helpful in some cases
23:14:02 Quit Dobson (Ping timeout: 252 seconds)
23:14:25fmlpixelma: my reasoning was that if the tuner does tuning in 0.05 steps (which it shouldn't), it's possible that it also does tuning in 0.0123 steps (or another bad number). SHould we also dsiplay that without cutting?
23:15:04fmlAnd the current code, i.e. the old FM screen showed frequency with only one digit.
23:15:37fmlpixelma: oh, sorry, I'm wrong! There were two
23:16:11fmlpixelma: but the first reason still stands! :-)
23:16:14kisakis there any problem with just letting the theme decide if it wants 1 digit or 2 int eh fm screen?
23:16:23pamauryfunman: your patch works, not surprising though
23:16:32kugelfml: ^
23:16:39fmlkisak: unnecessary complexity of the tag
23:16:41 Quit merbanan (Ping timeout: 260 seconds)
23:17:21kisakfml: I was thinking provide two digits and let the theme clip the second digit if it wants, that's all
23:17:48CIA-5New commit by funman (r26118): as3525: correctly detect usb status if we boot with USB plugged
23:17:57fmlkisak: how would you clip?
23:18:53kisakfml: I lack an answer to that ... I haven't seen enough of the code to wrap my head around it
23:19:54 Quit Xerion (Quit: )
23:21:12kugelpixelma: any results?
23:21:30pixelmafml: sorry, I can't take this argument seriously because 0.05 MHz steps seems to be the precision at it works now and is visible to me if I look at the presets list but which is hidden now in the FM screen. One way out is to make the tuner work at 0.1 MHz precision too at those regional setting but that's just not the way it is now
23:21:47pamaurysomeone needs to explain to me how the as353x patch for linux could work. Sometime the bitfield are msb->lsb and sometimes, it's the contrary ! How is this supposed to work oO
23:22:04pixelmakugel: with the inttypes it seems to continue compiling but there were warnings
23:23:09fmlpixelma: but what is that setting for then? (I mean the step.)
23:24:04pixelmaok, I have to check again
23:24:31kugelpixelma: paste them please
23:25:45fmlpixelma: I think if you change the frequency manually in single steps, i.e. without auto scanning/auto tuning, then this setting is taken into account. If you do auto tuning then the tuner might stop at an intermediate frequency.
23:26:50pixelmakugel: only one left actually
23:26:53pixelma/home/Marianne/SVN/rockbox/firmware/target/hosted/sdl/system-sdl.c:175: warniIn function `system_init'ng: implicit declaration of function `SDL_CreateThread' (In function `system_init')
23:27:26pixelmaerr... there's a c+p error in there
23:27:51 Quit bmbl (Quit: Bye!)
23:27:55pixelma/home/Marianne/SVN/rockbox/firmware/target/hosted/sdl/system-sdl.c:175: warning: implicit declaration of function `SDL_CreateThread' (In function `system_init')
23:28:11saratogakugel: when you get a chance could you review the pictureflow hotkey patch?
23:28:25saratogai believe you're the only one looking at pictureflow in the last year
23:28:31 Quit Rob2223 (Quit: Rob2223)
23:29:00 Join Rob2222 [0] (
23:29:09kugeli can have quick glance
23:29:23pixelmakugel: seems to be a working sim again at least
23:29:52saratogakugel: its FS #11270
23:31:09 Join stripwax [0] (
23:32:24kugelpixelma: strange error
23:32:36kugelor warning that is
23:34:41 Nick Ypsy is now known as YPSY (
23:34:58pamauryI'm really puzzled when I look at the linux code. Something like that:
23:34:59 Join fdinel [0] (
23:35:16pamaurya = b where a is a 1-bit field and b a 2-bit value that can either be 0, 1 or 2
23:35:24pamauryand then a test on the value of a !!!
23:36:26*linuxstb grumbles that JdGordon didn't bother with "svn cp" when splitting the radio code
23:40:41kugelpixelma: I don't think that warning can be fixed nicely, SDL_CreateThread should be in SDL.h which is included there
23:41:05kugelto me it seems like an sdl bug on your system actually
23:42:09CIA-5New commit by dave (r26119): Add missing svn properties for the new files created in r26109
23:43:09CIA-5New commit by kugel (r26120): Restore changes of r25809 that got lost in the target tree movement and fix simulator build on cygwin.
23:45:05 Quit GeekShadow (Quit: The cake is a lie !)
23:45:25Blue_Dudekugel: I'm getting the same warning...
23:45:30 Join Peter93 [0] (
23:46:04linuxstbkugel: "man SDL_CreateThread" says: #include "SDL.h" #include "SDL_thread.h" - which the code used to include IIUC
23:46:43 Quit Peter93 (Client Quit)
23:47:19Blue_DudeSDL_CreateThread is only defined in SDL_thread.h.
23:48:48kugeloh, I see
23:48:59Blue_DudeIt doesn't appear to be referenced in any other header file.
23:49:02*kugel wonders why it doesn't happen on linux
23:49:20Blue_DudeAnd SDL_thread.h isn't referenced in any other header file either.
23:49:26funmanUSB detection in mkamsboot works but I must wait ~100ms
23:49:37linuxstbBlue_Dude: On where? Cygwin?
23:49:38funmanssorgatem_: ping
23:49:49Blue_DudeI'm searching Cygwin, yes.
23:50:18linuxstbIn my SDL install (Linux), SDL.h is including SDL_thread.h
23:50:40linuxstbBut the manpage still says to include SDL_thread.h explicitly.
23:51:09 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
23:51:34kugellinuxstb: maybe they've changed it in a newer version?
23:51:36Blue_DudeIt probably won't hurt anything to include SDL_thread.h explicitly.
23:51:37linuxstbMaybe it's related to the SDL version - I have 1.2.14
23:51:39 Quit fml (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
23:51:48CIA-5New commit by kugel (r26121): Fix another warning which strangely only occurs on cygwin.
23:52:00***Saving seen data "./dancer.seen"
23:52:11linuxstbkugel: I would guess so - the SDL.h I have is including all other SDL .h files...
23:55:23Blue_DudeThe SDL files in Cygwin are dated Nov 2007. I don't know the version number. They're the latest in Cygwin too.
23:55:37linuxstbDo you have an SDL_version.h ?
23:55:57 Quit lpereira (Quit: Leaving.)
23:56:20linuxstbOr what does "sdl-config −−versoin" say? (if you have that)
23:56:52Blue_DudeYes to sdl_version: $Id: SDL_version.h,v 1.16 2005/04/17
23:57:17linuxstbIf you scroll down, you'll see three lines - SDL_MAJOR_VERSION etc
23:57:26Blue_Dudesdl-config reports 1.2.9
23:58:01CIA-5New commit by funman (r26122): Clip+: boot to OF if USB is connected ...
23:58:06 Quit ender` (Quit: Pets are better than children because: 11. If they get pregnant, you can sell their children.)
23:58:44pamauryfunman: do you think internal dma means the usb phy has a dma engine ?
23:58:46funmanssorgatem_: ^ this should work on fuzev2 too, get in touch with me when you can test a patch
23:58:58funmanpamaury: yes i think that's what ranma found on as3525

Previous day | Next day